AAEESS113300 th CCoonnvveenn ttiioonn PPrroo ggrraamm May 13 – 16, 2011 Novotel London West, London, UK

The AES has launched a new opportunity to recognize 1TNO Human Factors (retired), Soesterberg, student members who author technical papers. The Stu - The Netherlands dent Paper Award Competition is based on the preprint 2Embedded Acoustics, Delft, The Netherlands manuscripts accepted for the AES convention. A number of student-authored papers were nominated. This year, the Speech Transmission Index cele - The excellent quality of the submissions has made the brates its 40th anniversary. While the first mea - selection process both challenging and exhilarating. suring device built in the 1970s could barely fit The award-winning student paper will be honored dur - inside a car, inexpensive pocket-size STI mea - ing the Convention, and the student-authored manuscript suring solutions are now available to the world. will be considered for publication in a timely manner for Meanwhile, the STI method has continually the Journal of the Audio Engineering Society . evolved in order to deal with an increasing array Nominees for the Student Paper Award were required of measuring challenges. This paper investi - to meet the following qualifications: gates how the STI kept up with these challenges and analyzes possible room for further improve - (a) The paper was accepted for presentation at the ment. Also, a roadmap for further development AES 130th Convention. of the STI is proposed. (b) The first author was a student when the work was Convention Paper 8315 conducted and the manuscript prepared. (c) The student author’s affiliation listed in the manu - script is an accredited educational institution. 10:00 (d) The student will deliver the lecture or poster pre - sentation at the Convention. P1-2 Prosody Generation Module for Macedonian Text-to-Speech Synthesis— Branislav * * * * * Gerazov, Zoran Ivanovski , Faculty of Electrical Engineering and Information Technologies, The Winner of the 130th AES Convention Skopje, Macedonia Student Paper Award is: The paper presents a fully functional prosody Design of a Compact Cylindrical Loudspeaker Array generation module developed for Macedonian for Spatial Sound Reproduction — Mihailo text-to-speech (TTS) synthesis. The module is Kolundzija, Christof Faller , Martin Vetterli , based on research of prosody generation mod - Ecole Polytechnique Fédérale de Lausanne, ules in high-end TTS synthesis systems, previ - Lausanne, Switzerland ous prosody experiences in Macedonian TTS, Convention Paper 8336 as well as original research of prosody carried To be presented on Friday, May 13 in Session P4 through by the authors. The paper starts with an —Multichannel and Spatial Signal Processing, overview of the basic tasks, problems, and solu - Part 1 tions in prosody generation modules. Then it continues to give a detailed account of the work - * * * * * ings of the developed module. The module first segments the input speech into intonation phras - Friday, May 13 09:00 Room Saint Julien es and determines their intonation type. Next it Technical Committee Meeting on Transmission generates durations for each of the units that will and Broadcasting be used to synthesize the speech output. Then it determines the positions of the lexical stresses Session P1 Friday, May 13 09:30 – 12:30 and modifies these units’ durations. After deter - Room 4 mining the intonation phrase’s pitch accent loca - tion, it generates an adequate pitch contour and SPEECH AND HEARING calculates the pitch targets needed for unit modi - fication. The synthesis module uses this data to Chair: Bob Schulein , Asius Technologies LLC, generate prosody in the output speech. Generat - Longmont, CO, USA ed prosody patterns in the output speech are of satisfactory quality for arbitrary text input. The 09:30 presented results are of significant value for P1-1 The Evolution of the Speech Transmission Macedonian TTS and can be used for other Index— Herman J. M. Steeneken, 1,2 Sander J. under-represented languages. van Wijngaarden ,2 Jan A. Verhave 2 Convention Paper 8316 ¯

Audio Engineering Society 130th Convention Program, 2011 Spring 1 Technical Progra m 10:30 2University of Massachusetts, Amherst, MA, USA P1-3 The Influence of Transmission Channel on When a sound producing device such as insert the Admissibility of Speech Sample for earphones or a hearing aid is sealed in the ear Forensic Speaker Identification— Andrey canal, the fact that only a tiny segment of the Barinov , Speech Technology Center Ltd., sound wave can exist in this small volume at any St. Petersburg, Russia given instant, produces an oscillation of the stat - ic pressure in the ear canal. This effect can This paper is an extension and addition to greatly boost the SPL in the ear canal, especially papers previously published and presented dur - at low frequencies, a phenomena that we call ing the AES 39th Conference and AES 129th Trapped Volume Insertion Gain (TVIG). In this Convention, regarding voice sample quality study the TVIG has been found by numerical requirements and compensation for the influ - modeling as well as direct measurements using ence of transmission channels for further foren - a Zwislocki coupler and the ear of a human sub - sic or automatic speaker identification. In this ject, to be as much as 50 dB greater than sound paper we provide the analysis of different types pressures typically generated while listening to of transmission channels such as land line, sounds in an open environment. Even at moder - GSM, radio, and VoIP. We analyze the important ate listening volumes, the TVIG can increase the parameters of voice samples obtained from low frequency SPL in the ear canal to levels these channels and compare the influence of dif - where they produce excursions of the tympanic ferent channels on important speaker identifica - membrane that are 100 to 1000 times greater tion voice biometric features. At the end of the than in normal open-ear hearing. Additionally, paper, there are some conclusions provided the high SPL at low frequencies in the trapped regarding the circumstances under which the volume of the ear canal, can easily exceed the particular recording can be accepted/rejected for threshold necessary to trigger the Stapedius forensic speaker identification. reflex, a stiffing response of the middle ear, Convention Paper 8317 which reduces its sensitivity and may lead to audio fatigue. The addition of a compliant mem - 11:00 brane covered vent in the sound tube of an P1-4 Optimizing the Acoustic and Intelligibility insert ear tip was found to reduce the TVIG by Performance of Assistive Audio Systems up to 20 dB, such that the Stapedius reflex and Program Relay for the Hard of Hearing would likely not be triggered. and Other Users— Peter Mapp , Peter Mapp Convention Paper 8319 Associates, Colchester, UK 12:00 Around 10% of the general population suffer from a noticeable degree of hearing loss and P1-6 Can We Compare the Sound Quality of Noise would benefit from some form of hearing assis - Reduction between Hearing Aids? A Method tance or deaf aid. DDA legislation and require - to Level the Ground between Devices— Rolph ments mean that many more hearing assistive Houben, Inge Brons, Wouter A. Dreschler , systems are being installed—yet there is contin - Academic Medical Center, Amsterdam, uing evidence to suggest that many of these The Netherlands systems fail to perform adequately and provide This paper proposes the application of an equal - the benefit expected. This paper reports on the ization filter to remove unwanted differences in results of acoustic performance testing of a num - frequency response between recordings from ber of trial ALS systems. The use of STIPA, as a different hearing aids. The filter makes it possi - practical measure for assessing the potential ble to compare the perceptual effects (such as intelligibility of ALS systems, is discussed. The user preference) of a specific signal processing ALS microphone type, distance, and angular feature (e.g., noise reduction) between different location from the target acoustic source are hearing aids, without the dominant influence of shown to have a significant effect on the resul - differences in their frequency responses. Both tant potential intelligibility performance. The an objective quality measure (PESQ) and a lis - effects of typical ALS signal processing have tening experiment have shown that the filter was been investigated and are shown to have a able to “level the ground” between the devices small but significant effect on the STIPA result. included. The potential application of the inverse The requirements for a suitable acoustic test filter is to use it on recordings from hearing aids source to mimic a human talker are discussed so that we can directly compare the noise reduc - as is the need to adequately assess the effects tion between devices. This allows one to deter - of both reverberation and noise. The findings of mine if users prefer a certain noise reduction this paper are also relevant to the installation over another, which could lead to improved and testing of educational sound field systems rehabilitation of hearing impaired listeners. as well as boardroom and conference room Convention Paper 8320 systems. Convention Paper 8318 Session P2 Friday, May 13 09:30 – 12:30 Room 1 11:30 P1-5 Sound Reproduction within a Closed Ear LOUDSPEAKERS Canal:Acoustical and Physiological Effects 1,2 1 — Samuel Gido , Robert B. Schulein , Stephen Chair: John VanderKooy , University of Waterloo, 1 Ambrose Waterloo, Ontario, Canada & BW Group, 1 Asius Technologies LLC, Longmont, CO, USA Steyning, UK 2 Audio Engineering Society 130th Convention Program, 2011 Spring 09:30 In recent papers a lumped parameter model, which can simulate the impedance of conven- P2-1 Vibrations in the Loudspeaker Enclosure tional electro-dynamic transducers accurately, Evaluated by Hot Wire Anemometry has been presented. The new model includes and Laser Interferometry— 1 Danijel Djurek , frequency-dependent damping, which questions Nazif Demoli 2 1 traditional engineering practices in simulations AVAC (Alessandro Volta Applied Ceramics) of loudspeaker enclosures and, in particular, Laboratory for Nonlinear Dynamics, associated losses. In this paper the conse- Zlatar-Bistrica, Croatia 2 quences of frequency-dependent damping are Institute of Physics, Zagreb, Croatia evaluated to aid the development of simulations Vibration states of the loudspeaker enclosure and models of loudspeaker enclosures. were examined by the laser interferometry and Convention Paper 8324 hot wire anemometry. According to simple expressions, air pressure in the enclosure is inde - 11:30 pendent on air density. This statement has been tested by the use of gases indicated by very dif - P2-5 Com parison of M easurem ent M ethods for the ferent densities (air, He, SF6), and interferometric Equalization of Loudspeaker Panels Based data show a strong dependence of vibration on Bending W ave Radiation— Lars Hörchens, states of the enclosure walls on density, when dri - Diemer de Vries, Delft University of Technology, ving frequency increases from 500 Hz up to 1 Delft, The Netherlands kHz. The main reason for a deviation resides in the imaginary part of the Morse impedance exert - Spatially extended panel loudspeakers based on ed by the vibrating gas within the enclosure. bending wave radiation, such as distributed Convention Paper 8321 mode loudspeakers or multi-actuator panels, exhibit a complex radiation pattern with frequen- 10:00 cy-dependent directivity characteristics. This paper seeks to determine the kind, position, and P2-2 Losses and Coupling in Long Multi-Wire number of measurements needed to obtain an Loudspeaker Cables— Xavier Meynial, average radiation spectrum enabling efficient Guennolé Gapihan , Active Audio, Saint- equalization. To this end, three approaches are Herblain, France compared: wave field extrapolation of the panel surface normal velocity, extrapolation of a near- It is quite frequent in PA systems to use very long field pressure measurement, and actual in-situ loudspeaker cables. These PA systems some - measurements at a number of random positions times carry several audio signals in adjacent inside the listening space. The choice of the wires of the same cable, or several cables may be positions and the required number of measure- neighboring over very long distances such as sev - ments are discussed. Measurements taken on a eral hundreds of meters. This paper deals with multi-actuator panel are used to compare the dif- losses and coupling in these long cables. It is ferent approaches and present numerical shown that even though losses and coupling may results. introduce significant crosstalk between wires and Convention Paper 8325 affect the amplifier loads, it is possible to use very long loudspeaker cables to connect line arrays in PA systems. Coupling between cables is also 12:00 investigated, and strategies for reducing losses ite-Sized Baffles on M obile and coupling are presented. P2-6 The Effect of Fin Audio— 1 Convention Paper 8322 Device Personal Jordan Cheer, Stephen J. Elliott,1 Youngtae Kim,2 Jung-Woo Choi2 10:30 1University of Southampton, Southampton, UK P2-3 Subwoofers in Rooms: Modal Analysis for 2Samsung Electronics Co. Ltd., Korea Loudspeaker Placement— Juha Backman , To reduce the annoyance from the use of loud- Nokia Corporation, Espoo, Finland and Aalto speakers on mobile devices, previous work has University, Espoo, Finland investigated the use of acoustic contrast control Use of multiple subwoofers in a room is known to optimize the performance of small arrays of to help in reducing the variation of response both loudspeakers. These investigations have as a function of frequency and a function of assumed that the baffle dimensions are negligi- place, but simple geometry-based placement ble so that the loudspeakers are omnidirectional, rules guarantee good results only in symmetrical which is reasonable at low frequencies; howev- cases. The paper discusses the use of experi - er, in practice the effect of a finite-sized baffle on mental modal analysis and numerical optimiza - the optimized performance is important at higher tion based on modal behavior to determine the frequencies. This paper reports the results of optimal placement of single or multiple sub - using a finite-element model of a two-source ar- woofers in rooms with arbitrary geometry and ray, positioned on a mobile phone sized baffle, surface properties. to investigate the influence of the baffle on the Convention Paper 8323 predicted array performance. The baffle is shown to reduce the performance of the array at 11:00 frequencies greater than around 1 kHz, but then the directivity of the individual drivers enhances P2-4 Losses in Loudspeaker Enclosures— Claus performance at these higher frequencies. Futtrup , Scan Speak A/S, Videbæk, Denmark Convention Paper 8326 ¯

Audio Engineering Society 130th Convention Program, 2011 Spring 3 Technical Progra m Friday, May 13 10:00 Room Saint Julien 11:00 Technical Committee Meeting on Audio Recording and Mastering Systems P3-2 Detection of “Solo Intervals” in Multiple Microphone Multiple Source Audio Applications — Elias Kokkinis ,1 Joshua Reiss ,2 Workshop 1 Friday, May 13 John Mourjopoulos 1 10:30 – 12:30 Room 5 1University of Patras, Patras, Greece 2Queen Mary University of London, London, UK MUSIC AND SEMANTIC WEB In this paper a simple and effective method is Co-Chairs: David De Roure , University of Oxford proposed to detect time intervals where only a Yves Raimond , BBC single source is active (solo intervals) for multi - ple microphone, multiple source settings com - monly encountered in audio applications, such Panelists: David Bretherton , Southampton University as live sound reinforcement. The proposed Gregg Kellogg , Connected Media Experience method is based on the short term energy ratios Alexandre Passant , Seevl between all available microphone signals, and a Evan Stein , Decibel single threshold value is used to determine if and which source is solely active. The method is This workshop will provide an inter-disciplinary forum to computationally efficient and results indicate that explore and promote the combination of music and it is accurate and fairly robust with respect to Semantic Web technologies—specifically, to develop an reverberation time and amount of source understanding of how Semantic Web technologies can interference. contribute to the growth of music-related data on the Convention Paper 8328 Web, as well as applications of this data. Activities in this area are becoming well established, driven by projects in both research and industry, and they span personal 11:00 applications together with all aspects of the creation, management, discovery, delivery and analysis of musical P3-3 A Real-Time Sound Source Localization content. The ambitions of this inaugural workshop are to and Enhancement System Using Distributed bring together this emerging and energetic community, to Microphones— Amparo Marti, Maximo Cobos, share information and practice, and especially to articu - Jose J. Lopez , Universidad Politecnica Valencia, late the research agenda in Music and the Semantic Valencia, Spain Web. Through this we aim to facilitate innovation, inform The Steered Response Power - Phase Trans - Semantic Web research, and establish the basis and form (SRP-PHAT) algorithm has been shown to momentum for future events. be one of the most robust sound source localiza - tion approaches operating in noisy and reverber - Session P3 Friday, May 13 11:00 – 12:30 ant environments. A recently proposed modified Foyer SRP-PHAT algorithm has been shown to pro - vide robust localization performance in indoor POSTERS: SOUND FIELD ANALYSIS environments without the need for having a very fine spatial grid, thus reducing the computational 11:00 cost required in a practical implementation. Sound source localization methods are com - P3-1 Localization of Multiple Speech Sources monly employed in many sound processing Using Distributed Microphones— Maximo applications. In our case, we use the modified Cobos, Amparo Marti, Jose J. Lopez , SRP-PHAT functional for improving noisy Universidad Politecnica Valencia, Valencia, speech signals. The estimated position of the Spain speaker is used to calculate the time-delay for each microphone and then the speech is Source localization is an important task in many enhanced by aligning correctly the microphone speech processing systems. There are many signals. microphone array techniques intended to pro - Convention Paper 8329 vide accurate source localization, but their per - formance is severely affected by noise and reverberation. The Steered-Response Power 11:00 Phase Transform (SRP-PHAT) algorithm has been shown to perform very robustly in adverse P3-4 Binaural Moving Sound Source Localization acoustic environments; however, its computa - by Joint Estimation of ITD and ILD— Cheng tional cost can be an issue. Recently, the Zhou, Ruimin Hu, Weiping Tu, Xiaochen Wang, authors presented a modified version of the Li Gao , Wuhan University, Wuhan, Hubei, China SRP-PHAT algorithm that improves its perfor - mance without adding a significant cost. Howev - Spatial cues ITD and ILD that provide sound lo - er, the performance of the modified algorithm calization information play a very important role has only been analyzed in single source local - in the binaural localization system. The efficient ization tasks. This paper explores further the improvement of binaural moving sound source possibilities of this localization method by con - localization method by joint estimation of ITD sidering multiple speech sources simultaneously and ILD based on Doppler effect is investigated. active. Experiments considering different num - By removing Doppler effect influence, results ber of sources and acoustic environments are show that the proposed binaural moving sound presented using simulations and real data. source localization method achieves 0.3%(veloc - Convention Paper 8327 ity = 1m/s), 5.7%(velocity = 5m/s), and

4 Audio Engineering Society 130th Convention Program, 2011 Spring 10.5%(velocity = 10m/s) accuracy improvement mentation of an enhanced approach for a digital in silent conditions. The performance of our artificial reverberator. Starting from a preliminary method will be more effective as sound moves analysis of the mixing time, the selected impulse faster. response is decomposed in the time domain Convention Paper 8330 considering the early and late reflections. There - [Paper not presented but is available for purchase] fore, a short FIR is used to synthesize the first part of the impulse response, and a generalized 11:00 recursive structure is used to synthesize the late reflections, exploiting a minimization criterion in P3-5 Perceived Level of Late Reverberation in the cepstral domain. Several results are reported Speech and Music— Jouni Paulus ,1 Christian taking into consideration different real impulse Uhle ,1 Jürgen Herre 1,2 responses and comparing the results with those 1Fraunhofer Institute for Integrated Circuits IIS, obtained with previous techniques in terms of Erlangen, Germany computational complexity and reverberation 2International Audio Laboratories Erlangen, quality. Erlangen, Germany Convention Paper 8333 This paper presents experimental investigations on the perceived level of running reverberation in 11:00 various types of monophonic audio signals. The P3-8 Evaluation of Spatial Im pression Com paring design and results of three listening tests are dis - 2-Channel Stereo, 5-Channel Surround, cussed. The tests focus on the influence of the and 7-Channel Surround w ith Height input material, the direct-to-reverberation ratio, Channels for 3-D Im agery— Toru Kamekawa, 1 and the reverberation time using artificially gener - Atsushi Marui ,1 Toshikiko Date ,2 Masaaki Enatsu 3 ated impulse responses for simulating the late re - 1Tokyo University of the Arts, Tokyo, Japan verberation. Furthermore, a comparison between 2AVC Networks Company, Panasonic mono and stereo reverberation is conducted. It Corporation, Osaka, Japan can be observed that with equal mixing levels, the 3marimoRECORDS Inc., Tokyo, Japan input material and the shape of the reverberation tail have a prominent effect on the perceived lev - Three-dimensional (3-D) imagery is now widely el. The results suggest that mono and stereo spreading as one of the next visual formats for reverberation with identical reverberation times Blu-ray or other future media. Since more audio and mixing ratios are perceived as having equal channels are available with future media, the level regardless of the material. authors aim to find the suitable sound format for Convention Paper 8331 3-D imagery. A pairwise comparison test was carried out comparing combinations of 3-D and 11:00 2-D imagery with 2-channel stereo, 5-channel surround and 7-channel surround sound (5 P3-6 Reverberation Enhancement in a Modal channel surround plus 2 height channels) asking Sound Field— Hugh Hopper, David Thompson, better depth sense and better match between Keith Holland , University of Southampton, visual and audio images. The results show that Southampton, UK 3-D imagery with 7 channel surround gives the The reverberation time of a room can be highest sense of depth and match of visual and increased by using a reverberation enhance - audio images. ment system. These electronic systems have Convention Paper 8334 generally been installed in large rooms, where diffuse field assumptions are sufficiently accu - rate. Novel applications of the technology can be Tutorial 1 Friday, M ay 13 found by applying it to smaller spaces where iso - 11:00 – 12:30 Room 2 lated modal resonances will dominate at low fre - quency. An analysis of a multichannel feedback W HAT'S YO UR EQ IQ ? system within a rectangular enclosure is pre - sented. To assess the performance of this sys - Presenter: Alex Case , University of Massachusetts at tem, metrics are defined based on the spatial Lowell, Lowell, MA, USA and frequency variations of a diffuse field. These metrics are then used to optimize the parame - ters of the system using a genetic algorithm. It is Equalization—we reach for it track by track and mix by shown that optimization significantly improves mix, as often as any other effect. Easy enough, at first. the performance of the system. EQ becomes more intuitive when you have a deep Convention Paper 8332 understanding of the parameters, types, and technolo - gies used, plus deep knowledge of the spectral proper - 11:00 ties and signatures of the most common pop and rock P3-7 An Advanced Implementation of a Digital instruments. Alex Case shares his approach for applying Artificial Reverberator— Andrea Primavera, EQ and strategies for its use: fixing frequency problems, Stefania Cecchi, Laura Romoli, Paolo Peretti, fitting the spectral pieces together, enhancing flattering Francesco Piazza , Universita Politecnica delle features, and more. Marche, Ancona, Italy Reverberation is a well known effect particularly important for listening to recorded and live Friday, M ay 13 11:00 Room Saint Julien music. In this paper we propose a real imple - Technical Com m ittee M eeting on Autom otive Audio ¯

Audio Engineering Society 130th Convention Program, 2011 Spring 5 Technical Progra m Special Event Session P4 Friday, May 13 14:00 – 17:00 AWARDS PRESENTATION AND KEYNOTE ADDRESS Room 1 Friday, May 13, 12:45 – 13:30 Room 1 MULTICHANNEL AND SPATIAL SIGNAL PROCESSING, PART 1 Opening Remarks: • Executive Director Roger Furness • President Jim Kaiser Chair: Francis Rumsey , Logophon, Ltd., Witney, UK • Convention Chair Peter Mapp Program: • AES Awards Presentation 14:00 • Introduction of Keynote Speaker by Convention Chair Peter Mapp P4-1 3-D-Sound in Car Compartments Based on • Keynote Address by Trevor Cox Loudspeaker Reproduction Using Crosstalk Cancellation— Awards Presentation Andre Lundkvist, Arne Nykänen, Roger Johnsson , Luleå University of Technolo - Please join us as the AES presents Special Awards to gy, Luleå, Sweden those who have made outstanding contributions to the Society in such areas of research, scholarship, and pub - One way to enhance driving safety is to use lications, as well as other accomplishments that have signal sounds. Driver attention may further be contributed to the enhancement of our industry. The improved by placing sounds in a 3-D space, awardees are: using binaural synthesis. For correct loudspeak - Board of Governors Award er reproduction of binaural signals, crosstalk • Jan Berg between the channels needs to be cancelled • Kimio Hamasaki out. In this study, a crosstalk cancellation algo - • Toru Kamekawa rithm was developed and tested. The algorithm • Michael Kelly was applied in a car compartment, and three loudspeaker positions were compared. A listen - Fellowship Award ing test was performed to determine the sub - • Andrzej Brzoska jects’ ability to correctly localize sounds. It was • Christof Faller shown that loudspeakers placed behind the lis - • Masami “Sam” Toyoshima tener correctly reproduced sound sources in the back hemisphere. Loudspeakers located in front Bronze Medal Award and above the listener gave a high number of • C. Robin Caine front/back confusions for all angles. • Yoshizo “Steve” Sohma Convention Paper 8335

Keynote Speaker 14:30 This year’s Keynote Speaker is Trevor Cox . Trevor Cox P4-2 Design of a Compact Cylindrical is Professor of Acoustic Engineering at the University of Loudspeaker Array for Spatial Sound Salford, a Senior Media Fellow funded by the Engineer - Reproduction— Mihailo Kolundzija, 1 Christof ing and Physical Sciences Research Council, and Presi - Faller ,1 Martin Vetterli 1,2 dent of the Institute of Acoustics (IOA). 1Ecole Polytechnique Fédérale de Lausanne, One major strand of his research is room acoustics for Lausanne, Switzerland intelligible speech and quality music production and re - 2University of California at Berkeley, Berkeley, production. Cox’s diffuser designs can be found in rooms CA, USA around the world. He has co-authored a research book entitled Acoustic Absorbers and Diffusers . He was Building acoustic beamformers is a problem awarded the IOA’s Tyndall Medal in 2004. whose solution is hindered by the wide-band Cox has a long track record of communicating acoustic nature of the audible sound. In order to achieve engineering to the public and has been involved in en - a consistent directional response over a wide gagement projects worth over £1M. He was given the range of frequencies, a conventional acoustic IOA award for promoting acoustics to the public in 2009. beamformer needs a high number of discrete He has developed and presented science shows to loudspeakers and be large enough to achieve a 15,000 pupils including performing at the Royal Albert desired low-frequency performance. The Hall, Purcell Rooms, and the Royal Institution. Cox has acoustic beamformer design described in this presented fifteen documentaries for BBC radio including: paper uses measurement-based optimal beam - Life’s soundtrack, Save our Sounds, and Science vs the forming for loudspeakers located mounted on a Strad. His latest BBC radio 4 program due for broadcast rigid cylindrical baffle. Super-directional beam - in June is titled “Green Ears.” The title of Cox’s keynote forming enables achieving desired directivity speech is “Sounding Places, Past and Present: How with multiple loudspeakers at low frequencies. Spaces Affect the Sounds We Make and Hear.” High frequencies are reproduced with a single Architectural acoustics affect the perceived quality of loudspeaker, whose highly directional reproduc - live and recorded performances. Caves with ringing tion—due to the cylindrical baffle—matches the speleothem show that mankind has been exploiting design goals. In addition to the beamformer filter venues with distinctive acoustics since the upper Palae - design procedure, it is shown how such a loud - olithic era. This talk will explore unconventional places, speaker array can be used for spatial sound both ancient and modern, with remarkable acoustics: a reproduction. spherical room that allows you to whisper into one of your Convention Paper 8336 own ears; a water cistern with a reverberation time of over 15:00 forty seconds, and the whispering arches of Grand Central in New York. What do these spaces say about our hearing P4-3 A New Multichannel Microphone Technique abilities and the importance of sound to our lives? for Effective Perspective Control— Hyunkook

6 Audio Engineering Society 130th Convention Program, 2011 Spring Lee , University of Huddersfield, Huddersfield, UK Laurent Simon ,1,2 Russell Mason 1 1University of Surrey, Guildford, Surrey, UK This paper introduces a new multichannel micro - 2now with INRIA – Rennes, Bretagne Atlantique, phone technique that was designed to produce France multiple listener perspectives. A listener perspec - tive paradigm and the related spatial attributes are In previous studies, the localization accuracy also proposed for the evaluation of spatial quality and the spatial impression of 3-2 stereo micro - in acoustic recording and reproduction. The pro - phone arrays were discussed. These showed posed microphone array employs five coincident that 3-2 stereo cannot produce stable images to microphone pairs, which can be transformed into the side and to the rear of the listener. An octa - virtual microphones with different polar patterns gon loudspeaker array was therefore proposed. and directions depending on the mixing ratio. The Microphone array design for this loudspeaker results of interchannel correlation and informal configuration was studied in terms of localization subjective evaluations suggest that the proposed accuracy, locatedness, and image width. This technique is able to offer an effective control over paper describes an experiment conducted to various spatial attributes. evaluate the spaciousness of 10 different micro - Convention Paper 8337 phone arrays used in different acoustical envi - ronments. Spaciousness was analyzed as a 15:30 function of sound signal, acoustical environment, P4-4 Experimental Analysis of Spatial Properties and microphone array’s characteristics. It of the Sound Field Inside a Car Employing a showed that the height of the microphone array Spherical Microphone Array— Marco Binelli, and the original acoustical environment are the Andrea Venturi, Alberto Amendola, Angelo two variables that have the most influence on Farina , University of Parma, Parma, Italy the perceived spaciousness, but that micro - phone directivity and the position of sound A 32-capsule spherical microphone array was sources is also important. employed for analyzing the spatial properties of Convention Paper 8340 the sound field inside a car. Both the background noise and the sound generated by the car’s sound system were spatially analyzed, by superimposing false-color sound pressure level maps over a Session P5 Friday, May 13 14:00 – 17:30 panoramic 360x180 degree image, obtained with Room 4 a parabolic-mirror camera. The analysis of the noise field revealed the parts of the car body PRODUCTION AND BROADCAST where more noise is leaking in, providing guid - ance for better soundproofing. The analysis of the impulse responses generated by the loudspeak - Chair: Natanya Ford ers did show useful information on the reflection patterns, providing guidance for adding absorbent 14:00 material in selected locations and for optimizing position and orientation of loudspeakers. P5-1 Frequency Weighting and Ballistics for Convention Paper 8338 Program Loudness Modeling— Ian M. Dash ,1 Benjamin Smith ,2 Densil Cabrera 2 16:00 1Australian Broadcasting Corporation, Sydney, NSW, Australia P4-5 Improved ITU and Matrix Surround 2University of Sydney, Sydney, NSW, Australia Downmixing— Christof Faller ,1 Pieter Schillebeeckx 2 There has been both experimental and anecdo - 1Illusonic LLC, St-Sulpice, Swtizerland tal evidence that the low-frequency performance 2Soundfield Ltd., Wakefield, UK in the ITU-R Recommendation BS.1770 program loudness measurement algorithm could be Improvements to ITU and matrix surround down - improved. A listening test with an emphasis on mixing are proposed. The surround channels are low frequency content was therefore conducted. often mixed into the downmix with reduced gain An attempt was made to analyze the results in to prevent that the resulting stereo signal is over - octave bands by multiple regression, but larger ly ambient. A method is proposed that allows than expected variability precluded any useful control of the amount of ambience in the down - outcome from this method. A simpler regression mix signal independently of direct sound. In a analysis was therefore performed using several conventional matrix surround downmix, ambi - fixed weighting curves and asymmetric integra - ence from the surround channels appears tion with a range of time constants. Although the impaired (negatively correlated). To address this results largely support the present low frequency issue, a technique is proposed that separates weighting curve, they also indicate that asym - direct and ambient sound. Then, a matrix sur - metric integration provides better program loud - round downmix is only applied to the direct ness assessment than symmetric integration or sound, while ambient sound is treated like in an high level gating. ITU downmix. Convention Paper 8341 Convention Paper 8339 14:30 16:30 P5-2 Evaluation of Live Meter Ballistics for P4-6 Spaciousness Rating of 8-Channel Loudness Control— Scott Norcross ,1 Stereophony-Based Microphone Arrays— 2 1 Félix Poulin , Michel Lavoie ¯

Audio Engineering Society 130th Convention Program, 2011 Spring 7 Technical Progra m 1Communications Research Centre, Ottawa, are critical in mixdowns because most of them Ontario, Canada cannot be identified adequately. The newly sug - 2CBC/Radio-Canada, Montreal, Quebec, Canada gested model is to be used in automatic mix - down algorithms and as an evaluating measure The broadcast community is transitioning from in future development whenever masking sce - practices that revolved around audio peak nor - narios are to be described. malization to one where the focus is on loudness Convention Paper 8344 consistency. Central to this effort is the loudness measurement algorithm described in ITU-R BS.1770. To assist in the mixing of audio that 16:00 targets specific long-term loudness levels a need P5-5 The “Digital Solution”: The Answer to a Lot has been identified for some form of loudness- of Challenges within New Production based live audio metering. The EBU has Routines at Today’s Broadcasting Stations— proposed meter specifications to indicate Stephan Peus , Georg Neumann GmbH, Berlin, “momentary” and “short-term” loudness whose Germany ballistics are defined, respectively, by a 400 ms and a 3000 ms integration window. In parallel The introduction of networked production sys - with the EBU efforts, the CBC/Radio-Canada, in tems allows a very actual, topical, and efficient collaboration with the CRC, has been studying workflow especially within broadcast and TV since 2008 various loudness meter ballistics that applications. As a result production is moving would be suitable in a production environment. closer to editorial work. Sound editing such as This paper reports on a series of tests carried voice-over, etc., has to be done more and more out by the CBC/Radio-Canada where various by editors who naturally don’t have that special momentary meter ballistics were evaluated it knowledge as a sound engineer. Live recordings was found that an IIR-based meter ballistic with and interactive TV in future will call for further a 400 ms time-constant was preferred. steps to simplify the production processes and to Convention Paper 8342 enhance reliability. We will explain practical examples of challenges from daily production 15:00 routines, will give answers to solve the problems by using digital technology, and will show the P5-3 Adaptive Dynamics Enhancement— Martin effects on a more simple and reliable workflow. Walsh, Edward Stein, Jean-Marc Jot , DTS, Inc., Convention Paper 8345 Scotts Valley, CA, USA Modern recordings are being mastered with 16:30 more and more aggressive dynamic range com - P5-6 Automatic Mixing and Tracking of On-Pitch pression in an attempt to generate content that Football Action for Television Broadcasts— is louder than previous releases. This can often Robert G. Oldfield, Benjamin G. Shirley , lead to large discrepancies in perceived loud - University of Salford, Salford, UK ness between tracks that were mastered at different periods of recent history. A commonly For the television broadcast of football in proposed solution to this problem involves the Europe, the sound engineer will typically have use of loudness normalization. While such nor - an arrangement of 12 shotgun microphones malization techniques help to reduce discrepan - around the pitch to pick up on-pitch sounds such cies in loudness, they cannot resurrect dynamics as whistle blows, players talking, and ball kicks, that were removed due to extreme levels of etc. Typically, during a match, the sound engi - dynamic range compression. This paper outlines neer will increase and decrease the levels of a technique for restoring the dynamics of mod - these microphones manually in accordance with ern music by continuously monitoring transient where the action is on the pitch at a given time signal behavior together with the associated to prevent the final mix being awash with crowd dynamic range levels. When dynamic range noise. As part of the EU funded project, Fasci - compression is likely, transients are restored to natE, we have developed an automatic mixing levels that are more expected for the type of algorithm that intelligently seeks key events on material being played. the pitch and turns on the corresponding micro - Convention Paper 8343 phones, the algorithm picks out the key events and automatically tracks the action eliminating 15:30 the need for manual tracking. Convention Paper 8346 P5-4 Describing the Transparency of Mixdowns: The Masked-to-Unmasked-Ratio— Philipp 17:00 Aichinger ,1,2 Alois Sontacchi ,2 Berit Schneider-Stickler 1 P5-7 High Quality “Radio” Broadcasting over the 1Medical University of Vienna, Vienna, Austria Internet— David Errock , Wave Science 2University of Music and Performing Arts Graz, Technology Ltd., London, UK Graz, Austria Broadcasting audio over the internet has dra - In this paper a model that predicts the trans - matically improved the experience for listeners, parency of mixdowns is proposed. The masked- including removing the geographic boundaries to-unmasked-ratio relates the original loudness of terrestrial transmission, allowing playback on of an instrument to its loudness in the mix. In portable devices and time-shifting content. Inter - order to assess this new measure a listening test net audio has primarily increased choice with is conducted. It is shown that instruments with a traditional broadcasters now competing along - masked-to-unmasked-ratio of 10% or smaller side “internet only” music stations. Internet

8 Audio Engineering Society 130th Convention Program, 2011 Spring distribution can also increase the audio quality, lengthy consultation with composers, acousti - beyond that of terrestrial and satellite transmis - cians, electricians, and audio-visual specialists, sion channel constraints; unlike video stream - and lessons learned along the way might be ing, which is of lower perceived quality than useful for those interested in adapting their own traditional transmission. With the sustainable space for similar purposes. data capacity of internet connections increasing, Engineering Brief 2 lossless audio can be streamed live to the con - sumer. This paper discusses the issues relating 14:00 to these distribution techniques and the challenges that will be experienced by the EB1-3 Time Alignment of Subwoofers in Large PA broadcasters, the distribution network, receiver Systems— Natàlia Milán, Joan Amate , Master manufacturers, and listeners. Audio Convention Paper 8347 In common PA systems the frequency range is divided into different ranges that are reproduced using different cabinets (subwoofers for the bass Session EB1 Friday, May 13 14:00 – 15:30 range and top cabinets for the mid-high range). Foyer This means different locations and positions of the sound sources and therefore notches and DESIGN AND ASSESSMENT peaks in the crossover range. Time alignment is needed to adjust the arrival time of frequencies 14:00 in the crossover area. But it’s not only a matter of distance, since the crossover is modifying the EB1-1 Queen Mary's “Media and Arts Technology phase, too. In this brief, a downscaled model is Studios” Audio System Design— Martin used to show how to measure the phase differ - James Morrell, Christopher A. Harte, Joshua ence of a two-source PA system and correct it D. Reiss, Queen Mary University of London, using FFT measuring software. Two situations London, UK are treated: two sources with overlapped fre - quency response and two sources sharing The Media and Arts Technology Studios is a crossover frequency. new facility at Queen Mary University of London Engineering Brief 3 linking together a previous Listening Room, with our new rooms: Control Room, Performance Lab, and Plant Room. This engineering report [Engineering Brief #4 was withdrawn] discusses our design philosophy for our given brief to create a “world-class facility” for a space 14:00 that is a “blank canvas” for researchers. We detail considerations for making an audio system EB1-5 Sound System Documentation for that is simple for standard recording and play - Construction Projects— Thomas Knauss , back while having a tremendous amount of rout - Marshall-KMK Acoustics, Ltd., Chappaqua, NY, ing options for users to create unique projects USA between all the connected spaces, featuring two separate spatial audio reproduction systems. This engineering brief will provide sound system The result is a 96 kHz/24 bit MADI based system engineers, users, and operators with the oppor - using a multimode fiber optic network and dedi - tunity to become familiar with the information, cated wordclock throughout. processes, and documentation required to suc - Engineering Brief 1 cessfully coordinate the electrical, rough-in, and general construction requirements of a profes - sionally installed sound system with the needs of 14:00 architects, engineers, and contractors responsi - EB1-2 Revitalizing the Denis Arnold Hall for ble for construction projects. The brief will Multichannel Electroacoustic Sound include a categorized list of the 24 points of Diffusion and Recording — Duncan Williams , information that every architect, engineer, and University of Oxford, Oxford, UK construction professional needs to know about each and every sound system device to be The Denis Arnold Hall, the flagship lecturing and implemented on a construction or renovation performance space in the Faculty of Music, project. The presenter maintains 30 years of Oxford University, has recently been the benefi - audio experience including front of house engi - ciary of a complete refurbishment, including neer for national recording artists, large scale dedicated design and specification for the perfor - integration as a sound systems contractor to mance and diffusion of electroacoustic composi - complete professional design services for recital tion and multichannel sonic art. The new config - halls, performing arts centers, corporate, enter - uration includes acoustic absorption and tainment, and house of worship facilities. insulation, low frequency management, and Engineering Brief 5 eight flexible full range satellite loudspeakers. This diffusion system is complemented by a full 14:00 range of playback formats (from 1/4" reel-to-reel, to Blu-ray, with line level patching for discrete or EB1-6 A Free Database of Head-Related Impulse “stemmed” multichannel performance/playback), Response Measurements in the Horizontal as well as 16 small diaphragm capacitor micro - Plane with Multiple Distances— Hagen phones, hung in 8 stereo pairs from the ceiling Wierstorf, Matthias Geier, Alexander Raake, on an integrated winch system. The design and Sascha Spors , T-Labs, Technische Universität ¯ configuration of the space necessitated a Berlin, Berlin, Germany

Audio Engineering Society 130th Convention Program, 2011 Spring 9 Technical Progra m A freely available database of Head-Related Workshop 2 Friday, May 13 Impulse Response (HRIR) measurements is 14:00 – 16:00 Room 2 introduced. The impulse responses were mea - sured in an anechoic chamber using a KEMAR WHAT'S THE USE OF RECORDING? manikin at four different loudspeaker dis - tances—3 m, 2 m, 1 m, and 0.5 m—reaching Chair: Sean Davies , S.W. Davies Ltd., Aylesbury, UK from the far field to the near field. The loud - speaker was positioned at ear height and the manikin was rotated with a step motor in one Panelists: Ted Kendall , Independent Engineer, Wales, UK degree increments. For the 3 m distance addi - Robert Philip , Open University, Milton tional measurements have been carried out Keynes, UK where the torso stayed fixed and only the artifi - Gordon Reid , Cedar Audio Ltd., Cambridge, UK cial head was rotated. In addition to the raw As engineers we tend not to think too much about the impulse responses there are also data-sets eventual purpose of what we do, but this can influence available with a frequency response compensat - not only our own techniques, but can have ramifications ed for the use of several different headphones. far afield. A panel of specialists will consider recording as Engineering Brief 6 evidence in legal cases; as propaganda; as a true record of musical styles for different periods; as a way of 14:00 delivering drama; and as a means of transmitting secret intelligence. EB1-7 Loudness Measurement and Human Interpretation in Television Program Quality Student/Career Development Event Checking— Matthieu Parmentier , France THE CHANGING AUDIO PROFESSIONAL Television Friday, May 13, 14:00 – 15:00 Along the standardization of Loudness Measure - Room 5 ment (EBU R128 and ITU-R BS.1770 update), Melvyn Toms France Television introduced loudness in 2010 to Moderator: , JAMES (Joint Audio Media improve human interpretation during Quality Education Services) Checking processes. After the defined measure - Panelists: Samantha Bennett , University of ment itself, done by a machine respecting the Westminster, UK standard, studies have been conducted to fix Dennis Weinreich , APRS/UK Screen/JAMES objective limits according to the network's skills and viewer environments. By using the tools of - Over the previous 30 years, the technical and opera - fered by the EBU R128, we have introduced speci - tional demands placed upon working professionals in the fications for audio acceptance beyond the loud - audio industry have grown exponentially. Equally over ness target of –23 LUFS, already taken into this period the opportunities to learn and gain the neces - account. This work and results have been extend - sary experience to become accomplished by sitting and ed for common programs and short commercials working alongside established practitioners has fallen. in different ways. The measurement tools, devel - To a large extent, public and private education has taken oped according R128, have also been graphically on this role but has it been successful? developed to facilitate this additional reading. Audio production and technology courses attract a Engineering Brief 7 large intake of industry hopefuls but will they be “sitting and working alongside established practitioners” and 14:00 how can they be assured their course meets industry requirements? EB1-8 Do Young People Actually Care about the Quality of Their MP3s?— Ainslie Harris , Robert Friday, May 13 14:00 Room Saint Julien Gordon University, Aberdeen, Scotland, UK Technical Committee Meeting on Archiving, Restoration, and Digital Libraries Ten focus groups were conducted in 2010 in Toronto, New York, and northern England, with participants ranging in age from 15 to 32 years Friday, May 13 14:00 Room Mouton Cadet old. Participants were asked about their file qual - Standards Committee Meeting SC-02-02 on Digital ity preferences for downloading music from Input/Output Interfacing online services, in the context of their existing positive/negative experiences, and being able to Friday, May 13 15:00 Room Saint Julien design a new service that would offer them the Technical Committee Meeting on Human Factors in format(s) of their choice. This brief will outline Audio Systems some of the key qualitative findings, including: (1) Contrary to popular opinion, young people Tutorial 2 Friday, May 13 can tell the difference between a 128 k and 320 16:00 – 18:00 Room 5 k MP3, but there appears to be an age thresh - old, below which listeners either do not notice or FORENSIC AUDIO ENHANCEMENT do not care. (2) Listening context and environ - ment are important, and affect consumers’ file Eddy B. Brixen quality preferences, even when only considering Chair: , EBB Consult, Denmark what type of MP3 to download. (3) Consumers still find that they must weigh their personal Panelists: Andrey Barinov , Speech Technology Center quality preferences against other practical Ltd., Russia requirements. Robin P. How , Metropolitan Police, London, UK Engineering Brief 8 Gordon Reid , CEDAR Audio Ltd., Cambridge, UK

10 Audio Engineering Society 130th Convention Program, 2011 Spring Attendees of this tutorial should expect to learn more 16:30 about issues surrounding the enhancement of forensic P6-3 Capacitor “Sound” in Microphone Preamplifier audio recordings. Topics will include evidence handling DC Blocking and HPF Applications: and processing, problem-specific approaches, and the Comparing Measurements to Listening increasing enumeration of GSM modulated voice signals Tests— will be presented and discussed. The panel of presenters Robert-Eric Gaskell , McGill University, will include practitioners from law enforcement, software Montreal, Quebec, Canada manufacturers, and the scientific community. Before and The sonic effect of capacitors in various aspects after audio examples will be demonstrated focusing on of audio electronics design has long been dis - specific problems. cussed and speculated upon. Recent publica - tions have tested many of these theories through Friday, May 13 16:00 Room Saint Julien rigorous distortion measurements of a variety of Technical Committee Meeting on Audio capacitor types under several test conditions. for Telecommunications One particularly interesting result is a measur - able increase in 2nd harmonic distortion for elec - trolytic and PET type capacitors when a DC bias Session P6 Friday, May 13 16:30 – 18:00 is applied. This paper repeats these measure - Foyer ments for a set of capacitors commonly used in +48V blocking applications and high pass filters POSTERS: AUDIO EQUIPMENT in microphone preamplifier designs. These phys - ical measurements are then compared to the result of double blind listening tests in order to 16:30 examine the audibility of these capacitor distor - tions as well as explore their sonic effect on vari - P6-1 Study and Analysis of the Carrier Distortion ous program materials. Sources in PWM DCI-NPC Modulator— Vicent Convention Paper 8350 M. Sala, Luis Romeral , UPC-Universitat Politecnica de Catalunya, Terrassa, BCN, Spain 16:30 This paper studies and analyzes an analog P6-4 1W 104dBA SNR Filterless Fully-Digital Class PWM Modulator for a Half-Bridge DCI-NPC Am - D Audio Amplifier with EMI Reduction plifier (Diode Clamped Inverter – Neutral Point Technique— Rossella Bassoli, Carlo Crippa, Clamped) as one of the sources of distortion in Federico Guanziroli, Germano Nicollini , the switching amplification chain. The four types ST-Ericsson, Agrate Brianza, Monza Brianza, of error or distortion sources are studied and an - Italy alyzed: Carriers Offset Error (COE), Carriers Phase Error (CPE), Carriers Symmetry Error A 1W filter-less power DAC featuring 104 dBA (CSE), and Carriers Amplitude Error (CAE). This SNR and EMI spreading is presented. Pre-distor - paper concludes that the two major sources of tion algorithms are used to reduce harmonic dis - error or distortion in PWM modulation process tortion inherent to the employed modulation for DCI-NPC topology are the Amplitude (CAE) process, and an oversampling noise shaper allows and Offset (COE) Errors, the latter being largest reducing modulator clock speed to facilitate hard - ware implementation while keeping high-fidelity contributor to the total distortion. 2 Convention Paper 8348 quality. No analog circuits exist from I S interface [Paper not presented but is available for purchase] to speaker, leading to zero output offset and good efficiency even at medium/low power levels. 1.2V digital and 2.7V-5V output supplies are used. Active area is 0.94mm 2 in a 0.13micron CMOS 16:30 process. Total harmonic distortion at maximum level is about 0.2%. P6-2 Experimental Verification of an Electrostatic Convention Paper 8351 Transducer with a Partitioned Back Electrode— Libor Husník , Czech Technical 16:30 University in Prague, Prague, Czech Republic P6-5 Ultra-Low Power Audio Architecture for Electroacoustic transducers based on the con - Portable Devices— Kangeun Lee, Changyong denser principle have usually a planar back elec - Son, Dohyung Kim, Sihwa Lee , Samsung trode that is a one-piece entity. The previous Advanced Institute of Technology, Suwon, Korea work studied theoretically the possibility of mak - ing the back electrode partitioned. Such an Current portable devices demand not only higher arrangement can be used in a design of the performance but also lower power consumption. so-called digital loudspeaker, in which the sizes For the reason, this paper describes a low power of the partitioned electrode are in the ratio of the System-on-Chip (SoC) architecture that targets powers of 2, but this is only a special case. The playback multimedia format. To significantly aim of this work is to study performance of a reduce power consumption of the SoC, the sys - system modeling the electrostatic transducer tem exploits a DSP core to decode compressed with the partitioned back electrode by way of audio. It is a key technique that a pre-buffer measurement on an experimental sample. Vari - keeps compressed audio data. Therefore, the ous combinations of signals are applied on the DSP can independently playback audio data, partitioned electrode and the transducer and other systems including CPU, and DRAM is response is measured. able to be power off until all audio data remain - Convention Paper 8349 ing in the pre-buffer is exhausted by the DSP. ¯ Audio Engineering Society 130th Convention Program, 2011 Spring 11 Technical Progra m For seamless operation, we designed a new ker - UPC-Universitat Politecnica de Catalunya, nel driver that controls the DSP and CPU, and it Terrassa, BCN, Spain is embedded into Linux kernel sets of the Google’s Android 2.2. Energy efficiency is evalu - This paper presents a less studied source of effi - ated by using fourteen audio sequences encod - ciency losses in Multilevel Diode-Clamped- ed with mp3 of which format is 128 kbps stereo. Inverter or Neutral-Point-Converter (DCI-NPC) The experimental results show that the proposed Power Switching Amplifiers. Filter inductors gen - system required only 25% power of the conven - erally add another significant contribution to the tional DVFS algorithm and guaranteed Quality of total power loss in Power Switching Amplifiers Service (QoS) for mp3 playing. systems. This contribution is generally compara - Convention Paper 8352 ble to that of the switching power stage and it is important to obtain a reasonable accurate esti - mate of the losses. In this paper the four possi - 16:30 ble sources of losses in the demodulation filter P6-6 Design of a Passive DGRC Column are studied and analyzed and are defined by the Loudspeaker with Wave Front Synthesis— expressions to calculate the value of the losses. Xavier Meynial, Gilles Grégoire , Active Audio, Using these loss expressions, this paper ana - Saint-Herblain, France lyzes the contribution of each of the sources. This work finishes up presenting the results of The DGRC (Digital and Geometric Radiation simulation and the conclusions. Control) principle allows one to assign a large Convention Paper 8355 number of loudspeakers of a line array to a limit - [Paper not presented but is available for purchase] ed number of electronic channels. It was intro - duced in 2006 and is now used in digitally steerable column loudspeakers. In this paper we Student/Career Development Event propose a very simple and straightforward imple - OPENING AND STUDENT DELEGATE ASSEMBLY mentation of this principle in a passive column, MEETING – PART 1 where delays are approximated with all-pass passive circuits. As a result, the column is Friday, May 13, 16:30 – 18:00 placed vertically (not tilted) and radiates a wave Room 2 front that ensures high speech intelligibility and Chair: Daniel Deboy homogeneous SPL coverage. We present the design of this column, as well as experimental Vice Chair: Magdalena Plewa results. Finally, we discuss its advantages in The first Student Delegate Assembly (SDA) meeting is the terms of visual integration, acoustic perfor - official opening of the Convention’s student program and a mances, and cost effectiveness. great opportunity to meet with fellow students from all cor - Convention Paper 8353 ners of the world. This opening meeting of the Student Delegate Assembly will introduce new events and election 16:30 proceedings, announce candidates for the coming year’s P6-7 Design and Realization of a Reference Class election for the European and International Regions, Loudspeaker Panel for Wave Field Synthesis announce the finalists in the four new recording competi - — Stephan Mauer, Frank Melchior , IOSONO tion categories, and announce any upcoming events of the GmbH, Erfurt, Germany Convention. Students and student sections will be given the opportunity to introduce themselves and their activi - This paper describes the requirements, the ties, in order to stimulate international contacts. design, nd the realization of a reference loud - All students and educators are invited to participate in speaker panel for Wave Field Synthesis (WFS). this meeting. Election results and Recording Competition Beside the algorithm and the loudspeaker’s Awards will be given at the Student Delegate Assembly acoustical performance the quality of Wave Field Meeting–Part 2 on Monday, May 16. Synthesis is strongly dependent on the spatial aliasing frequency. Only below that frequency Friday, May 13 17:00 Room Mouton Cadet will the synthesized wave field be physically cor - Standards Committee Meeting SC-02-08 on Audio- rect. The spatial aliasing frequency is related to File Transfer and Exchange the distance of adjacent speakers in the loud - speaker array. To raise the aliasing frequency to Special Event 2.8 kHz a loudspeaker panel was designed with WELCOME RECEPTION AND PRESENTATION a tweeter spacing of 6 cm. The transducers, Friday, May 13, 18:00 – 19:30 electronics, and directivity were designed to Room 1 obtain an excellent sound quality and SPL cov - erage. Simulations regarding the influence of Fraunhofer IIS invites all visitors to the AES 130th Con - speaker spacing and wave field artifacts have vention to meet in a social atmosphere to catch up with been made. Measurement results of the panel friends and colleagues from the world of audio. There will are given. The overall system design will be be a short presentation of the “International Audio Labo - shown as an application example. ratories Erlangen—The New Center of Audio Research” Convention Paper 8354 followed by a Welcome Drinks Reception.

16:30 Special Event ORGAN RECITAL BY GRAHAM BLYTH P6-8 Study and Analysis of Demodulation Filter Friday, May 13, 20:00 – 21:30 Losses in DCI-NPC Multilevel Power Lincoln’s Inn Chapel, London Amplifiers— Vicent M. Sala, Luis Romeral , Graham Blyth’s traditional organ recital will be given on 12 Audio Engineering Society 130th Convention Program, 2011 Spring the new organ of the Lincoln’s Inn Chapel. The Chapel 09:00 has had a historic role in the life of the Inn. The present P7-1 User Driven, Local Model, Reclassification of building was consecrated on Ascension Day, 1623, and Drum Loop Audio Slices— Henry Lindsay- there are services every Sunday in law terms. Lincoln’s 1 2 1 Inn Chapel had to wait almost 200 years until the first or - Smith , Skot McDonald , Mark Sandler 1Queen Mary University of London, London, UK gan was installed in 1820. The modest organ of 1819-20 2 by Flight and Robson was replaced in 1856 by a three- FXpansion Audio Ltd., London, UK manual and pedal organ by William Hill, one of the finest We present a method for significantly improving organ builders in the country. This organ was rebuilt on the results of drum loop slice classification. An no less than nine occasions, most recently in 1969. onset detector is used to slice loops of percus - As the organ approached its 150th birthday it became sion only audio. Low level features are extracted clear that its increasing unreliability meant that a decision from the audio slices and the slices are classi - needed to be taken about its future. After exhaustive fied into one of seven percussion classes by a research it was concluded that so little of the original Hill previously trained PART decision table. This organ had remained unaltered that it was impossible to general classification algorithm shows only an restore it to any original condition. adequate performance. The user is then allowed After a lengthy tender process, a contract was signed to correct incorrect classifications. Each correct - with Kenneth Tickell in 2005 for a new three-manual ed classification is combined with a subset of the organ. It was assembled in the Northampton workshops original classifications and a nearest neighbor of Tickell in 2008-9, and arrived on site in July 2009. The algorithm reclassifies the remaining slices ac - case, of European oak, has been stained to match the cording to the corrected local model. The resul - woodwork of the Chapel, and the highly polished front tant algorithm converges on a 100% correct so - pipes are complemented by shades of lime wood carved lution, with nearly 40% fewer re-classifications to a foliage design based on the surviving example of than a non-assisted approach. early sixteenth-century mural painting displayed in the Convention Paper 8356 room that now forms the passage between Gatehouse Court and Hardwicke Building. The new organ was 09:30 installed by Kenneth Tickell in 2009-10. The recital will include Bach’s Great C major Prelude & P7-2 Kick-Drum Signal Acquisition, Isolation and Fugue, Franck’s Fantasie in A, two recent pieces Reinforcement Optimization in Live Sound— inspired by words of Sir John Betjeman by Dennis Wick - Adam J. Hill, 1 Malcolm O. J. Hawksford ,1 Adam ens, and a complete performance of Widor’s 5th Sym - P. Rosenthal, 2 Gary Gand 2 phony. Programs and maps will be available from the 1University of Essex, Colchester, Essex, UK Special Events Desk. 2Gand Concert Sound, Glenview, IL, USA Graham Blyth was born in 1948, began playing the piano aged 4 and received his early musical training as a A critical requirement for popular music in live- Junior Exhibitioner at Trinity College of Music in London, sound applications is the achievement of a England. Subsequently, at Bristol University, he took up robust kick-drum sound presented to the audi - conducting, performing Bach’s St. Matthew Passion ence and the drummer while simultaneously before he was 21. He holds diplomas in Organ Perfor - achieving a workable degree of acoustic isola - mance from the Royal College of Organists, The Royal tion for other on-stage musicians. Routinely a College of Music and Trinity College of Music. In the late transparent wall is placed in parallel to the kick- 1980s he renewed his studies with Sulemita Aronowsky drum heads to attenuate sound from the drum - for piano and Robert Munns for organ. He gives numer - mer’s monitor loudspeakers, although this can ous concerts each year, principally as organist and cause sound quality impairment from comb-filter pianist, but also as a conductor and harpsichord player. interference. Practical optimization techniques He made his international debut with an organ recital at are explored, embracing microphone selection St. Thomas Church, New York in 1993 and since then and placement (including multiple microphones has played in San Francisco (Grace Cathedral), Los in combination), isolation-wall location, drum- Angeles (Cathedral of Our Lady of Los Angeles), Ams - monitor electronic delay, and echo cancellation. terdam, Copenhagen, Munich (Liebfrauendom), Paris A system analysis is presented augmented by (Madeleine and St. Etienne du Mont) and Berlin. He has real-world measurements and relevant simula - lived in Wantage, Oxfordshire, since 1984 where he is tions using a bespoke Finite-Difference Time- currently Artistic Director of Wantage Chamber Concerts Domain (FDTD) algorithm. and Director of the Wantage Festival of Arts. Convention Paper 8357 He divides his time between being a designer of pro - fessional audio equipment (he is a co-founder and Tech - 10:00 nical Director of Soundcraft) and organ related activities. P7-3 Development of a Virtual Performance Studio In 2006 he was elected a Fellow of the Royal Society of with Application of Virtual Acoustic Arts in recognition of his work in product design relating Recording Methods— Iain Laird ,1 Damian to the performing arts. Murphy ,2 Paul Chapman ,1 Seb Jouan 3 1Glasgow School of Art, Glasgow, Scotland, UK 2 Session P7 Saturday, May 14 09:00 – 11:00 University of York, Heslington, York, UK 3 Room 1 Arup, Glasgow, Scotland, UK A Virtual Performance Studio (VPS) is a space that allows a musician to practice in a virtual ver - LIVE AND INTERACTIVE SOUND sion of a real performance space in order to acclimatize to the acoustic feedback received on Chair: Michael Kelly , Sony Computer Entertainment stage before physically performing there. Tradi - Europe, London, UK tional auralization techniques allow this by con - ¯

Audio Engineering Society 130th Convention Program, 2011 Spring 13 Technical Progra m volving the direct sound from the instrument with frequency band from 20 Hz to 1 kHz. This paper the appropriate impulse response on stage. In expands the previous work with measurements order to capture only the direct sound from the on several valve amplifiers, refines the model, instrument, a directional microphone is often and makes it applicable to macro signal levels used at small distances from the instrument. close to saturation of the transformer. Methods This can give rise to noticeable tonal distortion are also given to minimize this extra weakening due to proximity effect and spatial sampling of in transformers. the instrument’s directivity function. This work Convention Paper 8360 reports on the construction of a prototype VPS system and goes on to demonstrate how an 09:30 auralization can be significantly affected by the placement of the microphone around the instru - P8-2 Diaphonic Pump: A Sound-Activated ment, contributing to a reported “PA effect.” Alternating to Static Pressure Converter— Informal listening tests have suggested that Stephen D. Ambrose ,1 Robert Schulein ,1 there is a general preference for auralizations Samuel Gido 1,2 that process multiple microphones placed 1Asius Technologies LLC, Longmont, CO, USA around the instrument. 2University of Massachusetts, Amherst, MA, USA Convention Paper 8358 This paper discusses the operating principles and basic construction of a diaphonic pump, 10:30 which is a newly invented device for harvesting P7-4 Interactive Audio Realities: An Augmented / the energy inherent in sound waves and using it Mixed Reality Audio Game Prototype— Nikos to pump air, thereby pressurizing a vessel. Moustakas, Andreas Floros, Nicolas Grigoriou , Although this device is of general utility, the Ionian University, Corfu, Greece embodiment discussed in this paper is used to harvest sound energy from the speaker (bal - Audio-games represent a game alternative anced armature transducer) of a personal listen - based on audible feedback rather than on visual. ing device (headset or hearing aid), and use this They may benefit from parametric sound synthe - as a power source to inflate a bubble in the lis - sis and advanced audio technologies (i.e., aug - tener’s ear, thereby creating an acoustic seal. mented reality audio), in order to effectively real - The diaphonic pump utilizes a natural asymme - ize complex scenarios. In this paper a try in the flow pattern when fluid is alternatingly multiplayer game prototype is introduced that pushed and pulled, back and forth through a employs the concept of controlled mixed reality small orifice known as a “synthetic jet.” Sound in order to augment the sound environment of waves provide the alternating flow pattern each player. The prototype is realized as multi - across the synthetic jet orifice. Prototype dia - ple user audiovisual installations, which are phonic pumps were built, which attach to a back interconnected in order to communicate the sta - volume of a balanced armature transducer and tus of the selected control parameters in real- are small enough that the whole assembly, time. The prototype reveals significant relevance transducer, and pump, can fit in a human ear to the-well known on-line multiplayer games, canal. with its novelty originating from the fact that user Convention Paper 8361 interaction is realized in augmented reality audio [Paper presented by Bob Schulein] environments. Convention Paper 8359 10:00 P8-3 Scanning the Magnetic Field of Electro- Dynamical Transducers— Wolfgang Klippel , Session P8 Saturday, May 14 09:00 – 12:30 University of Technology, Dresden Germany, Room 4 Klippel GmbH, Dresden, Germany AUDIO EQUIPMENT The magnetic flux density in the magnetic gap and the geometry of the moving coil determine Chair: John Dawson the force factor Bl, which is an important para - meter of the electro-dynamical transducer. The paper presents a new measurement technique 09:00 for scanning the flux density B(z, φ) on a cylin - drical surface within and outside the magnetic P8-1 Signal Level and Frequency Dependent gap using a Hall sensor and robotics changing Losses Inside Audio Signal Transformers the position of the sensor versus vertical position and How to Prevent Those— Menno van der z and angle φ. The results derived from the Veen , ir. bureau Vanderveen, Zwolle, The scanning process reveal the real B field in the Netherlands gap considering the fringe field and irregularities In an earlier work (Convention Paper 7125) a in the magnetization, which may initiate a rock - model was presented that explains that low volt - ing mode and rubbing of the voice coil at higher age level audio signals are extra weakened amplitudes. Using the geometry of the coil the when they are fed through a transformer. This static force factor Bl(x, i=0) can be calculated as extra weakening is caused by the signal level a function of voice coil displacement x and and frequency-dependent inductance of the compared with the dynamic force factor B(x,i) transformer. Combining this extra weakening measured by a dynamic system identification with the threshold of hearing curves, showed techniques. Discrepancies between dynamic that noticeable loss of micro details occurs in the and static force factor characteristics are dis -

14 Audio Engineering Society 130th Convention Program, 2011 Spring cussed and conclusions for loudspeaker design environment. A number of recordings were per - and manufacturing are derived. formed in rooms with variable size and quality. Convention Paper 8362 Presets with beneficial frequency-dependent directivities are compared to settings with con - 10:30 stant directivity. It is discussed to what extend recordings can be further improved using the P8-4 Comparison of Anemometric Probe and plug-in. Tetrahedral Microphones for Sound Intensity Convention Paper 8365 Measurements— Giulio Cengarle, Toni Mateos , Fundació Barcelona Media, Barcelona, Spain 12:00 The measurement of sound intensity requires P8-7 Digital Microphones—What’s it All About?— the acquisition of sound pressure and acoustic John Willett , Circle Sound Services, Oxfordshire, velocity in a coincident position. Various trans - UK ducer topologies can be used to measure the acoustic velocity directly or indirectly. In this It’s now ten years since the first AES42 specifi - paper three transducers are compared: a pres - cation was published (AES42-2001) and the first sure-velocity anemometric probe and two tetra - AES42-compliant digital microphone came to the hedral B-Format microphones from different market. So this seems an opportune moment to manufacturers. The comparison has been car - look at AES42 digital microphones, their history, ried out in different fields, ranging from anechoic what they offer, the current market situation, and to diffuse, reverberant field conditions. The what the future may hold. analysis and comparison is based on intensimet - Convention Paper 8366 ric quantities such as the radiation index and the sound intensity vector. Strengths and limitations of the various approaches are reported, to suggest Session P9 Saturday, May 14 09:00 – 10:30 the preferred applications for each transducer. Foyer Convention Paper 8363 POSTERS: PERCEPTION AND EVALUATION 11:00 P8-5 Prediction of Perceived Width of Stereo 09:00 Microphone Setups— Hans Riekehof-Böhmer ,1 P9-1 The Effect of Loudness Overflow on Equal- Helmut Wittek 2 Loudness-Level Contours— 1HAW-Hamburg, Hamburg, Germany Andrew J. R. 2Schoeps Mikrofone, Karlsruhe, Germany Simpson, Joshua D. Reiss , Queen Mary Univer - sity of London, London, UK The diffuse-field correlation of the two signals gen - erated by a stereophonic microphone setup has This paper presents a formal derivation of the an effect on the perception of spatial width. A cor - Loudness Overflow Effect (LOE), which relation meter is often used to measure the corre - describes the impact of nonlinear distortion on lation coefficient. However, due to the frequency loudness. Computational analysis is then per - dependence of the correlation function, the corre - formed, comprised of two experiments involving lation coefficient is not an appropriate value for two compressive static nonlinearities and using predicting the perceived width when it comes to two well-known time-varying loudness models. time-delay stereophony. By using the newly The results characterize the nonlinearities in defined “Diffuse-Field-Image-Predictor” (DFI-Pre - terms of LOE as a function of frequency and of dictor) presented in this paper an attempt is made listening level in the case of 250-ms pure-tone to reliably predict perceived width. Listening tests stimuli, and in terms of the traditional equal-loud - show that the DFI-Predictor is fairly suitable for this ness-level contours. The analysis is then extend - task. The aim of the study is to compare the spa - ed to synthesized wind instruments for one of tial properties of different stereophonic microphone the nonlinearities. The effect of the nonlinearity techniques by one calculated value. on loudness as a function of musical note funda - Convention Paper 8364 mental frequency and listening level is described for various synthesized instruments. Convention Paper 8367 11:30 P8-6 Synthesis of Polar Patterns as a Function 09:00 of Frequency with a Twin Microphone: Audio Examples and Applications within the P9-2 Evaluating the Use of Audio Smartphone Apps for Higher Education— Creative Process of Music Mixing— Matthias Anne Nortcliffe, Kock ,1 Markus Kock ,2 Rainer Maillard, 3 Malte Andrew Middleton, Ben Woodcock , Sheffield Kob 1 Hallam University, Sheffield, UK 1Erich-Thienhaus-Institute, Detmold, Germany Digital audio technology has garnered interest in 2Leibniz Universität Hannover, Hannover, Education recently, being deployed by early Germany adopter academics to provide audio feedback. 3Emil-Berliner Studios, Berlin, Germany Students have also used it, gathering audio The directivity of a twin microphone can be notes on their personal devices to enhance their chosen by variable weighting of the two output learning. However, the sharing and distributing signals. In addition, the polar pattern can be of the recordings is time-consuming and requires adjusted as a function of frequency when con - separate technology. Smartphones with audio trolled with a VST Plug-in in a modern DAW apps are able to support recording and distribu - ¯

Audio Engineering Society 130th Convention Program, 2011 Spring 15 Technical Progra m tion/sharing of learning conversations more Microphone Techniques— Duncan Williams , effectively because of their additional customiz - University of Oxford (Wolfson College), Oxford, able and integrated functionality. This is attrac - UK tive to Education now that it is clear that smart - phones are becoming ubiquitous on campus. Choral recordings created on location were This paper describes an evaluation of audio evaluated perceptually to determine the nature apps for recording learning conversations by an of the variations in timbre that might be elicited academic and students and their experience in by the use of different stereo microphone tech - using smartphone audio apps to date. niques. Four stereo recordings were made Convention Paper 8368 simultaneously with coincident, near coincident, and spaced stereo microphone techniques. Lis - teners were invited to describe any perceived 09:00 changes through a verbal elicitation experiment, P9-3 A Study of Human Perception of Temporal informing an adjective “pool” of possible attribut - and Spectral Distortion Caused by es. These attributes were reduced in number to Subwoofer Arrays— Elena Shabalina ,1 six by verbal protocol analysis. The six remain - Janko Ramuscak ,2 Michael Vorländer 1 ing attributes were then scaled in a second lis - 1RWTH Aachen University, Aachen, Germany tening experiment. Mean and standard deviation 2d&b audiotechnik GmbH, Backnang, Germany values in the results suggested that there was variation in three timbral attributes. This illustrat - The key task for a sound reinforcement system ed that the manipulation of timbral attributes by is to provide an even sound pressure level distri - microphone technique, combined with perceptu - bution over the whole listening area with possi - al analysis, is possible. bly the same frequency response; and reduce Convention Paper 8371 radiation in wrong directions. For that a system should show a certain directivity, which for low 09:00 frequencies can be achieved only by using multi - ple sound sources, for example, placed in a row P9-6 Anchor Signals Validation for Two in front of the stage. This technique can help to Dimensions of a Four-Dimensional avoid strong interference and corresponding Perceptive Space— Yves Zango ,1,2,3 Régine space sound pressure level variations of the Le Bouquin-Jeannès ,2,3 Nathalie Costet ,2,3 conventional left/right setup of subwoofers. On Catherine Quinquis 1 the other hand, multiple sources cause changes 1Orange Labs Lannion Tech/Opera, Lannion of the impulse and frequency response of an Cedex, France, array. Listening tests showed that these 2INSERM U642, Rennes France changes are audible for experienced listeners. 3Université de Rennes, Rennes France Convention Paper 8369 The subjective assessment of speech and sound codecs requires anchor signals to ensure its reli - 09:00 ability. The reference system currently used is P9-4 Evaluation of the Psychoacoustic Perception Modulated Noise Reference Unit (MNRU), which of Geometric Acoustic Modeling-Based simulates only quantization noise. Now, the new Auralization — Aglaia Foteinou, Damian T. generations of codecs present other impair - Murphy , University of York, Heslington, York, UK ments. In this study we consider speech quality as a multidimensional phenomenon and use The subjective evaluation of the auralization of a dimensional reduction techniques to project simulated acoustic, with a view to establishing codecs’ impairments in a four-dimensional the success or otherwise of the results obtained, space, each axis of the perceptive space corre - is usually best achieved using listening tests sponding to one of them. A verbalization test comparing the virtual environment with the actu - allowed characterizing two of these dimensions al measured space. As existing modeling meth - by the following attributes: “muffle” and “back - ods still need to be improved, it is of critical ground noise.” Anchor signals were designed for importance to focus on the human perception of these two dimensions, and a statistical analysis acoustic of the given space, rather than optimiz - allowed validating the accuracy of at least one of ing room acoustic parameters based only on these signals. objective measures. This paper uses a much Convention Paper 8372 simplified representation of a space with the re - sulting computer model giving the capability to 09:00 control all of the examined acoustic or simulation parameters independently. A 3-D shoebox P9-7 Auditory Distance Perception: Criteria shape room is created and a variety of factors and Listening Room— Jean-Christophe are changed every time in order to investigate Messonnier ,1 Alban Moraud 2 their relevance and influence on human percep - 1Conservatoire de Paris CNSMDP, Paris, France tion. These results are obtained from listening 2Altia, Paris, France tests, and conclusions for the psychoacoustic perception of such a space are given. This paper is the result of a series of listening Convention Paper 8370 experiments carried out to investigate the corre - lation between auditory distance and two criteria: the ratio of direct to reverberant sound energy 09:00 and the clarity C80. In the first section of this P9-5 A Comparative Perceptual Evaluation of the paper we will determine which of the two criteria Timbral Variations in Choral Location is more efficient. The second section compares Recordings Created by Four Common Stereo the values of these criteria when the same signal 16 Audio Engineering Society 130th Convention Program, 2011 Spring is played on a well damped control room loud - Cleveland Clinic, Cleveland, OH, USA speaker system and when it is played on a Michael Santucci , Sensaphonics Hearing domestic stereophonic loudspeaker system. A Conservation, Chicago, IL, USA second series of listening experiments shows how the auditory distance is perceived in both cases . Tinnitus is a common yet poorly understood disorder Convention Paper 8373 where sound is perceived in the absence of an external source. Significant sound exposure, with or without hear - ing loss, is the most common risk factor. Tinnitus can be 09:00 debilitating and can impair quality of life. Anxiety, depres - P9-8 Subjective Comparison between Stereo sion, and sleep disorders are potential consequences. and Binaural Processing from B-Format Most importantly for those in the audio industry, it can Ambisonic Raw Audio Material— Fábio W. significantly impair auditory perception. Sousa , University of York, York, North Yorkshire, This tutorial will focus on methods in managing tinnitus UK in the life of an audio professional. Background informa - tion will be provided regarding the basic concept of tinni - Using audio recorded in Ambisonic B-format tus, pertinent anatomy and physiology, audiologic from a sound field microphone and processed parameters of tinnitus, and an overview of current through both stereo and binaural tools, a subjec - research. Suggestions for identifying and mitigating high tive comparison is made. Hearing tests were risk behaviors will be covered. Elements of medical and performed taking into account personal experi - audiologic evaluations of tinnitus will also reviewed. ence and preference, as well as some spatial attributes concepts defined in previous works. Student/Career Development Event Aiming to evaluate the real effectiveness of bin - EDUCATION FORUM PANEL aural processing, this paper considers the possi - Saturday, May 14, 09:00 – 10:45 bility of distributing contemporary music, original - Room 5 ly developed for Ambisonic reproduction, through either conventional stereo or a method Moderator: Alex Case , University of Massachusetts at of high fidelity spatial processing directed specif - Lowell, Lowell, MA, USA ically to reproduction through headphone sys - tems, based on binaural technology. The sound Teaching the Teachers—A Round Table Discussion images represented in both binaural and stereo Among Audio Educators processing are examined. Spatial attributes like While audio itself—in all her disciplines—advances at wideness, depth, naturalness, and presence are breakneck speed, the educators supporting it must make evaluated. equivalent progress. AES conventions are reliable cata - Convention Paper 8374 lysts for earnest discussions among audio educators. However the convention, rich with so many activities, always seems to end too soon. Curriculum, personnel, Workshop 3 Saturday, May 14 and facilities must offer both time-proven fundamentals 09:00 – 11:00 Room 3 and cutting edge innovations. It happens in multiple modes: the classroom, the studio, the lab, online, cam - PANNING FOR MULTICHANNEL LOUDSPEAKER pus committee meetings, and through industry relation - SYSTEMS ships. We've all found solutions here, and nuggets of wisdom there; we wrestle with challenges and unknowns Chair: Ville Pulkki , Aalto University School of elsewhere. Join this discussion as we seek to define and Science and Technology, Aalto, Finland prioritize the key issues facing educators and create a vision for the most effective way to address them in fu - ture AES activities—through conventions, conferences, Panelists: Filippo Fazi , ISVR, UK online interactions, and more. Share your ideas for the Matthias Frank , IEM, Austria most essential forms of research and the best ways to Florian Keiler , Technicolor DE, Germany present the results: publications, tutorials, workshops, Amplitude panning is the most used method to position and other collaborations. What are the topics educators virtual sources over layouts where the number of loud - need to discuss, and what is the best format for sharing speakers is between two and about forty. The method is our advancements? AES provides the essential commu - really simple, it provides a nice spatial effect, and does nity for sharing and learning among audio educators. not color the sound prominently. This workshop reviews Help us design the next steps for increasing our produc - the working principle and psychoacoustic facts of ampli - tivity, accelerating our innovation, enriching our cama - tude panning for stereophony and for multichannel lay - raderie, and enhancing our quality as educators. outs. Panelists will describe some recent improvement suggestions to amplitude panning, which target some Saturday, May 14 09:00 Room Saint Julien shortcomings of amplitude panning in spatial accuracy. A Technical Committee Meeting on Audio Forensics lively discussion is expected on the pros and cons of such processing. Student/Career Development Event STUDENT SCIENCE SPOT Tutorial 3 Saturday, May 14 Saturday, May 14 through Monday, May 16 09:00 – 11:00 Room 2 Exhibits Area Check out the Student Science Spot for student designs MANAGING TINNITUS AS A WORKING and projects! See the young creative geniuses demon - AUDIO PROFESSIONAL strate their knowledge in the flesh. Learn about other stu - dent events or just hang and meet some new audio Presenters: Neil Cherian , Neurological Institute, friends. This is an amazing opportunity to share your ¯

Audio Engineering Society 130th Convention Program, 2011 Spring 17 Technical Progra m hard work with convention participants! Hope to see you diameter dependent change of cut-off frequency. all there! http://www.aes.org/ students/blog/ Convention Paper 8376 2011/3/130th-student-science-spot-call 12:00 Saturday, May 14 10:00 Room Saint Julien Technical Committee Meeting on Microphones P10-3 Automatic Recognition of Events in Audio and Applications Data Using Supercomputer Cluster— Kuba Lopatka, Andrzej Czyzewski, Henryk Krawczyk , Gdansk University of Technology, Gdansk, Session P10 Saturday, May 14 11:00 – 13:00 Poland Room 1 Dangerous events’ automatic recognition by AUDIO CONTENT MANGEMENT audio analysis employing parallel processing on a supercomputer cluster is described in the Chair: Jamie Angus, University of Salford, Salford, paper. Sound files recorded by microphones op - Greater Manchester, UK erating in a security surveillance system are processed by a sound event detection and clas - sification algorithm. Because of the large amount 11:00 of data, parallel computation is employed to P10-1 A Comprehensive and Modular Framework speed up the analysis. The sound file recorded for Audio Content Extraction, Aimed at by the surveillance system is divided into chunks Research, Pedagogy, and Digital Library and processed by separate threads or process - Management— Olivier Lartillot , University es. Several strategies for such parallel computa - of Jyväskylä, Jyväskylä, Finland tion are introduced and discussed. Results obtained in tests using a supercomputer cluster We present a framework for audio analysis and are presented. the extraction of low-level features, mid-level Convention Paper 8377 structures, and high-level concepts, altogether studied as a fully interwoven complex system. 12:30 Composite operations are constructed via an intu - itive programming language on top of Matlab. P10-4 Using Support Vector Machines for Datasets of any size can be processed thanks to Automatic Mood Tracking In Audio Music— implicit memory management mechanisms. The Renato Panda, Rui Pedro Paiva , University data structure enables a tight articulation between of Coimbra, Coimbra, Portugal signal and symbolic layers in a unified framework. The resulting technology can be used as a peda - In this paper we propose a solution for automatic gogical tool for the understanding of audio, mood tracking in audio music, based on super - speech, and musical processes and concepts, vised learning and classification. To this end, and for content-based discovery of digital various music clips with a duration of 25 sec - libraries. Other applications includes intelligent onds, previously annotated with arousal and browsing and structuring of digital library, informa - valence (AV) values, were used to train several tion retrieval, and the design of content-based models. These models were used to predict audio interfaces. quadrants of the Thayer’s taxonomy and AV val - Convention Paper 8375 ues, of small segments from full songs, revealing the mood changes over time. The system accu - racy was measured by calculating the matching 11:30 ratio between predicted results and full song an - P10-2 Selected Playback Problems of Historical notations performed by volunteers. Different Grooved Media— Nadja Wallaszkovits, 1 Franz combinations of audio features, frameworks, and Lechleitner ,1 Heinrich Pichler 2 other parameters were tested, resulting in an 1Phonogrammarchiv Austrian Academy of accuracy of 56.3% and showing there is still Sciences, Vienna, Austria much room for improvement. 2Audio Consultant, Vienna, Austria Convention Paper 8378 The paper discusses some selected playback problems of the replay and high quality archival Workshop 4 Saturday, May 14 transfer of historical grooved media, like cylin - 11:00 – 13:00 Room 3 ders, instantaneous discs and early coarse groove records. The topics outline the problems of noise reduction and the compensation of the PRODUCTION FOR UPCOMING SPATIAL horizontal tracking angle by means of stereo AUDIO SYSTEMS playback and modification of the sum and differ - ential signals. A comparison between existing Chair: Frank Melchior , IOSONO, Erfurt, Germany noise reduction methods and analog as well as digital phase and group delay compensation Panelists: Marc Emerit , Orange Labs, France methods is given and discussed. Finally possible Kimio Hamasaki , NHK, Tokyo, Japan compensation methods for the change in noise Jeff Levison , IOSONO GmbH, USA spectrum caused by the groove velocity Jörn Loviscach , University of Applied Sciences, decrease at inner diameters with early discs are Bielefeld, Bielefeld, Germany outlined. The authors propose a radius equaliza - Wieslaw Woszcyk , McGill University, Montreal, tion in the digital domain by using a digital high Quebec, Canada pass filter without group delay distortion, using Gregor Zielinsky , Sennheiser, Hannover, Germany

18 Audio Engineering Society 122nd Convention Program, 2007 Spring Driven by the force of 3-D pictures the demand for spa - passage of a master to manufacture is out of step with tial audio experiences is increasing. Several systems any QC that takes place. This can and has resulted in have been proposed just adding more channels around sometimes major and costly mistakes. the auditorium. While the integration of new channels Our discussion will focus on the challenges posed by seems to be simple from a workflow perspective, prob - an emerging new shape of the industry where artists, lems arise if the content needs to be adapted to different small labels, and producers are dealing directly with setups in different venues. On the other hand, and driven manufacturers. Furthermore, these problems are also by new spatial audio reproduction methods, a paradigm common to downloaded products. The incidences that shift toward object-based audio production is currently have raised the issue of QC during the manufacturing under discussion in the community. This workshop stage of CDs can only be exacerbated if consumer deliv - brings together the experience and tools in production for ery moves toward high quality downloads. new multichannel formats and new spatial audio repro - The growing tendency to third party agents arranging duction methods in general. Examples of related produc - manufacture and distribution also raise issues in respect of tions and installations are given. delivery formats to manufacturers and aggregators and with the upsurge in delivery via FTP we need to establish Tutorial 4 Saturday, May 14 some good practice for checking that what goes to manu - 11:00 – 13:00 Room 5 facture is correct and most especially who is responsible. This discussion launches the Mastering Section of Music IN-EAR MONITORING—PAST, PRESENT, Producers Guild which will be headed by Ray Staff. They AND EMERGING DEVELOPMENTS propose to outline what quality control is and where the responsibilities lie and to publish a specification that will be posted on the MPG website so that anyone involved in the Stephen Ambrose Presenter: , Asius Technologies process can refer to for good practice and guidance. In the 1970s, sound engineer and performer Stephen Ambrose began pioneering work experimenting with the Student/Career Development Event use of sound isolating earphones as a better means of EDUCATION AND CAREER/JOB FAIR hearing himself and others during live performances. Saturday, May 14, 11:30 – 13:00 This form of monitoring was in sharp contrast to the norm Foyer at the time, which consisted of an array of dedicated on-stage loudspeakers. Unlike with loudspeakers, per - Institutions offering studies in audio (from short courses to formers with isolating earphones could hear their own graduate degrees) will be represented in a “table top” ses - unique mix, separate from others. As an additional bene - sion. Information on each school’s respective programs fit of in-ear monitoring, stage monitors are no longer pre - will be made available through displays and academic sent to leak output into the house sound system or guidance. There is no charge for schools to participate. recording mixes. Early work of Ambrose, with a wide The Career Fair will feature several companies from the range of well-known performers, established the basics exhibit floor. All attendees of the Convention, students of in-ear monitoring but was challenged by technical bar - and professionals alike, are welcome to come talk with riers and acceptance issues. Today, however, IEM representatives from the companies and find out more (In-Ear-Monitoring) is in widespread use in many product about job and internship opportunities in the audio indus - forms. This tutorial, punctuated with a variety of demon - try. Bring your resume! Admission is free and open to all strations, will cover key developments responsible for the Convention attendees. growing acceptance of IEM systems. Additionally, new improvements in the areas of sound isolation, comfort, Saturday, May 14 12:00 Room Saint Julien occlusion effect suppression, and reduction of hearing Technical Committee Meeting on Spatial Audio fatigue, most of which are not yet commercially available, will be discussed. Saturday, May 14 12:00 Room Mouton Cadet Standards Committee Meeting SC-02-12 on Audio Saturday, May 14 11:00 Room Saint Julien Applications of Networks Technical Committee Meeting on Coding of Audio Signals Saturday, May 14 13:00 Room Saint Julien Technical Committee Meeting on Studio Practices Special Event and Production AES/MPG EVENT: WHERE DOES THE BUCK STOP? QC IN THE MANUFACTURING CHAIN Session P11 Saturday, May 14 14:00 – 17:30 Saturday, May 14, 11:15 – 13:00 Room 1 Room 2 Moderator: Tony Platt , MPG Director ROOM ACOUSTICS Panelists: Ray Staff , Award-winning chief mastering engineer, Air Mastering Chair: Diemer de Vries , Delft University of Pieter Stenekes , Founder of Sonoris Audio Technology, Delft, The Netherlands Engineering who are responsible for the DDP delivery software 14:00 We are all human and can make mistakes. It is also said P11-1 DTS Multichannel Audio Playback System: that we do not recognize our own mistakes. In the video Characterization and Correction— Zoran and broadcast world it is standard for the client to pay for Fejzo ,1 James Johnston 2 an independent QC check at the studio and again at the 1DTS, Inc., Calabasas, CA, USA broadcaster's facility. With vinyl it has always been nor - 2DTS, Inc., Kirkland, WA, USA mal to have a test pressing. All these procedures have been bypassed in CD mastering, and, in general, the Audio playback system correction methods are ¯

Audio Engineering Society 130th Convention Program, 2011 Spring 19 Technical Progra m now commonplace in audio-video receivers. One audio engineer’s ability to effectively design, con - goal of these systems is to correct deviation of figure, and evaluate their sound systems. loudspeaker/ room frequency response from Convention Paper 8381 some desired target curve. Unfortunately this correction may be inappropriate outside of a 15:30 small area around the microphone location, and averaged measurements may provide unwanted P11-4 What’s Wrong with Scattering Theory?— Ian timbre shifts. Some room correction algorithms M. Dash, Fergus R. Fricke , University of Sydney, capture the room responses at multiple locations Sydney, NSW, Australia and combine them to obtain a representative The theory of long wave scattering from the side response that is used for frequency correction. of a cylinder originated with Rayleigh and was We will present a loudspeaker/room correction extended to a general solution by Morse. Both system that attempts to achieve perceptually solutions are based on an angular harmonic appropriate frequency correction in a wide listen - series expansion. This model is conceptually ing area by using a closely spaced non-coinci - flawed. A number of physical paradoxes inherent dent multi-microphone array placed in a single in the model are outlined and discussed. location in the room. By use of special probe sig - Convention Paper 8382 nals, this is achieved within a short measure - ment period. 16:00 Convention Paper 8379 P11-5 New Proposals for the Calibration of Sound 14:30 in Cinema Rooms— Philip Newell ,1 Keith Holland ,2 Julius Newell ,3 Branko Neskov 4 P11-2 Evaluating the Auralization of Performance 1Acoustics Consultant, Moaña, Spain Spaces and Its Effect on Singing 2ISVR, University of Southampton, Performance— Judith Brereton, Damian T. Southampton, UK Murphy, David M. Howard , University of York, 3Electroacoustics Engineer, Lisbon, Portugal Heslington, York, UK 4Loudness Films, Lisbon, Portugal Musicians alter their performance according to The current practices for calibrating cinema the acoustic environment in which they perform, rooms date back to the early 1970s. Much has but as yet a thorough parametric investigation of been learned since then about the perception of the effect of room acoustics on musical perfor - sound, and measurement techniques have mance has not yet been achieved. A sufficiently advanced greatly. Evidence has been growing “realistic” synthesized Room Impulse Response that the present degree of room-to-room com - (RIR) will facilitate such a study, since this will patibility leaves much to be desired, and com - allow the investigator greater control and knowl - plaints about loudness and intelligibility problems edge of the room acoustic parameters involved. persist. This paper looks at reassessing the This paper reports the results of an experiment whole process of the loudspeaker and room cali - to evaluate a virtual acoustic space through the bration from a modern perspective. performance, interview, and audio analysis of Convention Paper 8383 the performance of a solo singer. Simulations of [Paper presented by Keith Holland] the same performance space using synthesized RIRs and measured RIRs were compared. In 16:30 general, singers who took part in the trial could distinguish between the two simulations and rat - P11-6 Some Preliminary Comparisons between ed the measured RIR simulation more highly in the Diffusion Equation Model and Room- terms of warmth and reverberance. Acoustic Rendering Equation in Complex Convention Paper 8380 Scenarios— Juan M. Navarro ,1 Jose Escolano ,2 Jose J. López 3 15:00 1San Antonio’s Catholic University of Murcia, Guadalupe, Spain P11-3 Enhancing the Configuration and Design 2University of Jaén, Linares, Spain of Sound Systems through Simulation— Fred - 3Universidad Politecnica de Valencia, Valencia, erick Otten, Richard Foss , Rhodes Spain University, Grahamstown, South Africa Recently, a model named acoustic radiative Audio Engineers are required to design and transfer equation has been proposed as a gen - deploy large multichannel sound systems that eral theory to expand geometrical room acoustic meet a set of requirements and use networking modeling algorithms. This room acoustic model - technologies such as Firewire and Ethernet. ing technique establishes the basis of two Bandwidth utilization and latency need to be recently proposed algorithms, the acoustic diffu - considered. Network Simulation can be used to sion equation model and the room acoustic accurately model a network and return such rendering equation. This paper presents some information. This paper discusses a software comparisons of room-acoustic parameters in-situ system that has been developed to create a sim - measurements with prediction values from both ulation of a network using the AES-X170 proto - methods in a real complex shape room in order col for command and control. This system shows to clarify advantages and limitations of both information about bandwidth and latency and is methods. Moreover, the memory requirements able to detect problems with parameter relation - and computation time have been evaluated. ships. It also provides the ability to perform offline Convention Paper 8384 editing. These features significantly enhance the

20 Audio Engineering Society 130th Convention Program, 2011 Spring 17:00 designed where room properties and reproduc - tion techniques were varied in a way that P11-7 Spatial Room Impulse Responses with allowed evaluation of noise and reverberation a Hybrid Modeling Method— Alex Southern, suppression based on speech intelligibility mea - Samuel Siltanen, Lauri Savioja , Aalto University, surements. Artificial head recordings were com - Aalto, Finland pared to in-ear recordings and real life listening. The synthesis of an arbitrary enclosure room im - Artificial head recordings were found to be pulse response (RIR) may be performed using equivalent to real life listening. The speech intel - acoustic modeling. A number of acoustic model - ligibility for in-ear recordings surpassed real life ing methods have been proposed previously listening. A possible explanation may be inaccu - each with their own advantages and limitations. rate equalization. The equalization is critical for This paper is concerned with mixing the RIRs correct reproduction of binaural cues. The proce - from different modeling methods to synthesize a dure used is convenient for validation of the hybrid RIR. Low frequencies are modeled using performance of recording and reproduction the finite difference time domain method equipment intended for sound quality studies. (FDTD), high frequencies are treated with geo - Convention Paper 8387 metric methods. A practical implementation for forming a hybrid RIR is discussed and further demonstrated in the context of a 2nd order B- 15:00 Format spatial encode of the modeled sound P12-3 Perceptually Robust Headphone Equalization field. The paper discusses the considerations for Binaural Reproduction— Bruno Masiero, and limitations of forming such hybrid RIRs us - Janina Fels , RWTH Aachen University, Aachen, ing wave-based and geometric-based methods. Germany Convention Paper 8385 Headphones must always be adequately equal - ized when used for reproducing binaural signals if they are to deliver high perceptual plausibility. However, the transfer function between head - Session P12 Saturday, May 14 14:00 – 17:30 phones and ear drums (HpTF) varies quite heav - Room 4 ily with the headphone fitting for high frequen - cies, thus even small displacements of the BINAURAL SOUND headphone after equalization will lead to irregu - larities in the resulting frequency response. Keeping in mind that irregularities in the form of Bozena Kostek Chair: , Gdansk University of peaks are more disturbing than equivalent val - Technology, Gdansk, Poland leys, a new method for designing headphone equalization filters is proposed where not the 14:00 average but an upper variance limit of many measured HpTFs is inverted. Such a filter yields P12-1 Comparison of Speech Intelligibility in perceptually robust equalization since the equal - Artificial Head and Jecklin Disc Recordings— ized frequency response will, with high chance, Roger Johnsson, Arne Nykänen , Luleå differ from the ideal response only by the pres - University of Technology, Luleå, Sweden ence of valleys in the high frequency range. Convention Paper 8388 Binaural recordings are often done using artifi - cial heads but can also be done with a Jecklin disc. In this study an experiment was designed 15:30 that allowed evaluation of noise and reverbera - tion suppression based on speech intelligibility P12-4 Prediction of Perceived Elevation Using measurements. Recordings of a voice and Multiple Pseudo-Binaural Microphones— disturbing noise were done in a reverberant en - Tommy Ashby, Russell Mason, Tim Brookes , vironment using one artificial head and four University of Surrey, Guildford, Surrey, UK Jecklin discs of various sizes. A listening experi - ment using headphones was conducted to deter - Computational auditory models that predict the per - mine the speech intelligibility in the recordings ceived location of sound sources in terms of and in a real life situation. It was found that there azimuth are already available, yet little has been was no significant difference in the speech intel - done to predict perceived elevation. Interaural time ligibility between the artificial head and Jecklin and level differences, the primary cues in horizontal disc with a diameter of 36 cm. localization, do not resolve source elevation, result - Convention Paper 8386 ing in the “Cone of Confusion.” In natural listening, listeners can make head movements to resolve such confusion. To mimic the dynamic cues provid - 14:30 ed by head movements, a multiple microphone sphere was created, and a hearing model was P12-2 A Comparison of Speech Intelligibility for developed to predict source elevation from the sig - In-Ear and Artificial Head Recordings— Arne nals captured by the sphere. The prototype sphere Nykänen, Roger Johnsson , Luleå University of and hearing model proved effective in both hori - Technology, Luleå, Sweden zontal and vertical localization. The next stage of this research will be to rigorously test a more physi - Good binaural reproductions should allow the lis - ologically accurate capture device. tener to suppress noise and reverberation as Convention Paper 8389 when listening in real life. An experiment was ¯

Audio Engineering Society 130th Convention Program, 2011 Spring 21 Technical Progra m 16:00 paper presents a method for extrapolation of measured HRTF datasets from the source dis - P12-5 BRTF (Body Related Transfer Function) tance used in the measurements to other source and Whole-Body Vibration Reproduction distances. The method applies the concept of Systems— M. Ercan Altinsoy, Sebastian local Wave Field Synthesis to compute extrapo - Merchel , Dresden University of Technology, lated HRTFs for almost arbitrary source Dresden Germany positions with high accuracy. The method is If binaural recorded signals are played back via computationally efficient and numerically stable. headphones, the transfer characteristic of the Convention Paper 8392 reproduction system has to be compensated for. Unfortunately, the transfer characteristic depends not only on the transducer itself, but Workshop 5 Saturday, May 14 also on mounting conditions and individual prop - 14:00 – 15:00 Room 3 erties of the respective ear. This is similar with reproduction systems for whole-body vibrations. AURO 3D—HIGH QUALITY AND LOW LATENCY PCM The transfer characteristic depends to a great ENCODING AND DATA REDUCTION extent on the individual body properties, e.g., weight or body mass index. In this study body Wilfried Van Baelen related transfer functions of 60 subjects are Chair: , Galaxy Studios, Belgium measured using an electrodynamic excitation Auro-3D is first and foremost a family of surround for - system. In addition anthropometric data of the mats including height information, starting with 4 over - subjects are collected. This paper reviews the head channels (and loudspeakers) and leading up to existing whole-body vibration reproduction sys - 13.1. The obvious use for such enlarged arrays is in the tems and discusses the importance of the indi - cinema, but the 9.1 Auro-3D solution is also applicable to vidual transfer functions for whole-body vibration the home theater, as the additional speakers are exactly reproduction. above the existing ones (L+R front, L+R surround). Be - Convention Paper 8390 sides being a family of formats, also a codec has been developed that can fold down 9.1 to 5.1 or 5.1 to 2.0 16:30 purely in the PCM domain, with ultra-low latency decod - ing (1ms). The man behind Auro-3D, Wilfried Van Baelen P12-6 HRTF-Enabled Microphone Array for Binaural from Galaxy Studios in Belgium, will give an introduction Synthesis— Malcolm O. J. Hawksford , to the Auro-3D codec, talk about latest developments, University of Essex, Colchester, Essex, UK and play some examples using the codec. A synthesis technique incorporating a circular phased-array microphone is described where Special Event the horizontal polar response is matched to an AES/APRS/DV247 EVENT: TALKBACK-PRO 1 arbitrary set of head-related transfer functions ACOUSTIC TREATMENT FOR SMALL SPACES (HRTFs). The array can emulate the function of Saturday, May 14, 14:00 – 15:45 an artificial listener but without the need to Room 2 embed physical anatomical features. Design Malcolm Atkin techniques are described based upon polar Moderator: response equalization computed over a discrete Panelists: TBA frequency space that together with dynamic coefficient processing enables spatial image As 3-D visuals hit the screens, there are more opportuni - manipulation and bespoke multi-listener environ - ties to provide surround-sound mixes. How can small ments with individual head tracking. A method of rooms best be set-up to mix in surround and if mixing/ 2-D spatial filtering is described to scale the control room space is terminally limited, can non- number of microphone signals. Applications acoustic solutions offer successful fixes? This perennial include binaural recording optimally matched to problem has plagued aspiring producers and production an arbitrary number of listeners, distributed gam - rooms since audio began it’s “democratizing” journey. ing, teleconferencing, multi-user interactive virtu - Hear studio acoustics experts and technologists al reality, and remote surveillance. debate and demonstrate the best way to exploit small Convention Paper 8391 production spaces for both conventional stereo and sur - round-sound mixes. 17:00 Saturday, May 14 14:00 Room Mouton Cadet P12-7 Interpolation and Range Extrapolation of Standards Committee Meeting SC-05-02 on Audio Head-Related Transfer Functions Using Connectors Virtual Local Wave Field Synthesis— Sascha Spors, Hagen Wierstorf, Jens Ahrens , Deutsche Student/Career Development Event Telekom Laboratories, Technische Universität STUDENT RECORDING CRITIQUES Berlin, Berlin, Germany Saturday, May 14, 15:00 – 16:00 Virtual environments that are based on binaural Sunday, May 15, 10:00 – 11:00 sound reproduction require datasets of head- Monday, May 16, 11:00 – 12:00 related transfer functions (HRTFs). Ideally, these Room Muscadet HRTFs are available for every possible position Moderator: Ian Corbett , Kansas City Kansas of a virtual sound source. However, in order to Community College, Kansas City, KS, USA reduce measurement efforts, such datasets are typically only available for various source direc - Following the success of these events at the recent San tions but only for one or very few distances. This Francisco convention—STUDENTS— bring a track to be

22 Audio Engineering Society 130th Convention Program, 2011 Spring critiqued in London! record production as an academic subject within the uni - These are non-competitive sessions, the idea being versity system relates to the nuts and bolts of doing the that you get “another” set of ears to listen to your work job. What kind of approaches are researchers engaging and make some constructive suggestions. Students are in? What do producers think about these types of encouraged to bring in their stereo or surround projects research and the emergence of practice led research in for feedback and comments from a panel and audience. this field? You will be able to sign-up for time slots at the first SDA meeting, on a first come, first served basis. Students who are finalists in the Recording Competition are Saturday, May 14 15:30 Room Mouton Cadet excluded from participating in this event to allow the non- Standards Committee Meeting SC-05-05 finalists an opportunity for feedback on their hard work. on Grounding and EMC Practices Bring your stereo or surround work on CD, DVD, or hard disc as clearly-labeled .wav files. The Student Recording Critiques are generously sponsored by PMC. Workshop 8 Saturday, May 14 16:00 – 18:00 Room 2 Saturday, May 14 15:00 Room Saint Julien Technical Committee Meeting on Perception HIGH RESOLUTION AUDIO PUBLISHING and Subjective Evaluation of Audio Chair: Stefan Bock , MSM Studios, Germany Workshop 6 Saturday, May 14 15:15 – 16:15 Room 3 Panelists: Morten Lindberg , 2L, Norway Darcy Proper , Wisseloord Studios, BOUNCE TO APP The Netherlands Tokuyama Takeshi , Imagica, Japan Neil Wilkes , Opus Productions Chair: Michael Hlatky , University of Applied Sciences Yunichi Yoshio , Pioneer, Japan Bremen, , Bremen, Germany Panelists: Jörn Loviscach , University of Applied Sciences, There are a great deal of formats suitable for high resolution Bielefeld, Germany audio delivery. Which ones will succeed? What challenges Martin MacMillian , Bounce Mobile do they need to overcome to become a standard? This pan - Martin Roth , RjDJ el will present a complete panorama of the available options, reviewing their advantages and disadvantages. Recorded music used to be a lean-back experience, but mobile devices have changed the game: Applications Saturday, May 14 16:00 Room Saint Julien such as interactive 360-degree music videos begin to Technical Committee Meeting Advisory Group leverage the nonclassical man-machine interfaces of on Regulations these devices—cameras, accelerometers, and GPS receivers—and employ their always-connectedness to access social music recommendation sites or music Session P13 Saturday, May 14 16:30 – 18:00 analysis and synthesis Web services. So far, the sound - Foyer tracks of these apps have been produced from the final mixes. Can there be a better integration of audio produc - tion and the development of interactive music applica - POSTERS: PRODUCTION AND BROADCAST tions? What would that mean in terms of workflow and user interfaces? The issues that arise in this context are 16:30 markedly different from those arising in the production of full-scale video games. P13-1 A Comparison of Kanun Recording Techniques as They Relate to Turkish Makam Workshop 7 Saturday, May 14 Music Perception — Can Karadogan , Istanbul 15:15 – 16:15 Room 5 Technical University, Istanbul, Turkey This paper presents a quality comparison of ART OF RECORD PRODUCTION microphone techniques applied on the kanun, a prominent traditional instrument of Turkish Co-Chairs: Katia Isakoff Makam music. Disregarding the effects of pre- Simon Zagorski-Thomas amplifier color, A/D converter, compression, Panelists: Haydn Bendall , Engineer/Music Producer equalization, mixing, and mastering, only the Steve D’Agostino , Artist/Music Producer studio recording step of music production is tak - Paschall de Paor , University of Glamorgan en into focus. Microphone techniques were Tony Platt , MPG and JAMES applied with varying placements and microphone types, and doing so, original Turkish Makam mu - This year will see the seventh international Art of Record sic etudes were recorded. Using short excerpts Production (ARP) Conference in San Francisco and the of these etudes, a survey comparing microphone re-launch of the ARP Journal . The Association for the techniques and placements as well as micro - Study of the Art of Record Production aims to provide an phone types was prepared. Subjects were cho - international and interdisciplinary organization for pro - sen from kanun players, sound engineers, and moting the study of the production of recorded music that non-musicians who showed different perspec - involves both academics and industry professionals. This tives, preferences, and descriptions to the sound workshop brings together the two directors of ARP with samples of various microphone techniques. top record producers to discuss how the development of Convention Paper 8393 ¯

Audio Engineering Society 130th Convention Program, 2011 Spring 23 Technical Progra m 16:30 tening tests were performed to assess usability and performance of the system, and the test pro - P13-2 Objective Measurement of Produced Music cedure and results are reported. Quality Using Inter-Band Dynamic Convention Paper 8396 Relationship Analysis— Steven Fenton ,1 Bruno Fazenda ,2 Jonathan Wakefield 1 16:30 1University of Huddersfield, Huddersfield, UK 2University of Salford, Salford, UK P13-5 The Quintessence of a Waveform: Focus and Context for Audio Track Displays— This paper describes and evaluates an objective Jörn measurement that grades the quality of a com - Loviscach , Fachhochschule Bielefeld (University plex musical signal. The authors have previously of Applied Sciences), Bielefeld, Germany identified a potential correlation between inter- Oscilloscope-style waveform plots offer great band dynamics and the subjective quality of pro - insight into the properties of the audio signal. duced music excerpts. This paper describes the However, their use is impeded by the huge previously presented Inter-Band Relationship spread of timescales extending from fractions of (IBR) descriptor and extends this work by testing a millisecond to several hours. Hence, waveform with real-world music excerpts and a greater plots often require zooming in and out. This number of listening subjects. A high degree of paper introduces a graphical representation correlation is observed between the Mean Sub - through a synthesized quintessential waveform ject Scores (MSS) and the objective IBR that shows the spectrum of the traditional wave - descriptor suggesting it could be used as an ad - form plot but does so at a much larger timescale. ditional model output variable (MOV) to describe The quintessential waveform can reveal details produced music quality. The method lends itself about single periods at zoom levels where a reg - to real-time implementation and therefore can be ular waveform plot only indicates the signal’s exploited within mixing, mastering, and monitor - envelope. Compression renders the enormous ing tools. ranges of frequencies and amplitudes more Convention Paper 8394 legible. Convention Paper 8397

16:30 16:30 P13-3 Evaluation of a New Algorithm for Automatic P13-6 Spatial Audio Processing for Interactive TV Hum Detection in Audio Recordings— Services— 1 1 1 2 Johann-Markus Batke, Jens Spille, Matthias Brandt , Thorsten Schmidt , 1 1 2 1 Holger Kropp, Stefan Abeling , Ben Shirley, Joerg Bitzer 2 1 Rob G. Oldfield Jade University of Applied Sciences, 1Technicolor, Research, and Innovation, Oldenburg, Germany 2 Hannover, Germany Cube-Tec International, Bremen, Germany 2University of Salford, Salford, UK In this paper an evaluation of a recently pub - FascinatE is a European funded project that lished hum detection algorithm for audio signals aims at developing a system to allow end users is presented. To determine the performance of to interactively navigate around a video panora - the method, large amounts of artificially generat - ma showing a live event, with the accompanying ed and real-world audio data, containing a vari - audio automatically changing to match the se - ety of music and speech recordings, are lected view. The audiovisual content will be processed by the algorithm. By comparing the adapted to the users particular kind of device, detection results with manually determined covering anything from a mobile handset ground truth data, several error measures are equipped with headphones to an immersive computed: hit and false alarm rates, frequency panoramic display connected with large loud - deviation of the hum frequency estimation, offset speaker setup. We describe how to handle audio of detected start and stop times, and the accura - content in the FascinatE context, covering sim - cy of the SNR estimation. ple stereo through to spatial sound fields. This Convention Paper 8395 will be performed by a mixture of Higher Order Ambisonics and Wave Field Synthesis. Our pa - 16:30 per focuses on the greatest challenges for both techniques when capturing, transmitting, and P13-4 Interactive Mixing Using Wii Controller— rendering the audio scene. Rod Selfridge, Joshua Reiss , Queen Mary Convention Paper 8398 University of London, London, UK 16:30 This paper describes the design, construction, and analysis of an interactive gesture-controlled P13-7 Wireless High Definition Multichannel audio mixing system by the means of a wireless Streaming Audio Network Technology Based video game controller. The concept is based on on the IEEE 802.11 Standards— Seppo Nikkila, the idea that the mixing engineer can step away Tom Lindeman, Valentin Manea , ANT – from the mixing desk and become part of the Advanced Network Technologies Oy, Helsinki, performance of the piece of audio. The system Finland allows full, live control of gains, stereo panning, equalization, dynamic range compression, and a A novel approach for the wireless distribution of variety of other effects for multichannel audio. uncompressed real-time multichannel streaming The system and its implementation are audio is presented. The technology is based on described in detail. Subjective evaluation and lis - the IEEE 802.11 Point Coordination Function,

24 Audio Engineering Society 130th Convention Program, 2011 Spring Contention Free Medium Access Control with Saturday, May 14 17:00 Room Saint Julien size-optimized multicast frames. The implemen - Technical Committee Meeting on Electro Magnetic tation supports eight independent audio streams Compatibility with simultaneous 24-bit audio samples at the rate of 192 kHz. Frame length alignment algo - Saturday, May 14 17:00 Room Mouton Cadet rithm is developed for smooth, low jitter flow. An Standards Committee Meeting IEC 60268-3 ad-hoc audio specific Forward Error Correction scheme meeting and a low system latency inter-channel synchro - nization method are described. The clock drift is Special Event solved by a sample stuffing/stripping algorithm. OPEN HOUSE OF THE TECHNICAL COUNCIL Implemented hardware and software structures AND THE RICHARD C. HEYSER MEMORIAL are presented and the technology is compared LECTURE with other wireless audio networking concepts. Saturday, May 14, 18:30 – 20:00 Emerging multichannel content formats are Room 1 briefly reviewed. Convention Paper 8399 Lecturer: Karlheinz Brandenburg The Heyser Series is an endowment for lectures by emi - 16:30 nent individuals with outstanding reputations in audio engi - P13-8 Gestures to Operate DAW Software— Wincent neering and its related fields. The series is featured twice Balin ,1 Jörn Loviscach 2 annually at both the United States and European AES 1Universität Oldenburg, Oldenburg, Germany; Conventions. Established in May 1999, The Richard C. 2Fachhochschule Bielefeld (University of Applied Heyser Memorial Lecture honors the memory of Richard Sciences, Bielefeld, Germany Heyser, a scientist at the Jet Propulsion Laboratory, who was awarded nine patents in audio and communication There is a noticeable absence of gestures—be techniques and was widely known for his ability to clearly they mouse-based or (multi-)touch-based—in present new and complex technical ideas. Heyser was mainstream digital audio workstation (DAW) soft - also an AES governor and AES Silver Medal recipient. ware. As an example for such a gesture consid - The Richard C. Heyser distinguished lecturer for the er a clockwise O drawn with the finger to 130th AES Convention is Karlheinz Bandenburg. Bran - increase a value of a parameter. The increasing denburg has been a driving force behind some of today’s availability of devices such as smartphones, most innovative digital audio technology, notably the tablet computers, and touchscreen displays rais - MP3 and MPEG audio standards. The research results es the question in how far audio software can of his dissertation are the basis of MPEG-1 Layer 3 benefit from gestures. We describe design (MP3), MPEG-2 Advanced Audio Coding (AAC), and strategies to create a consistent set of gesture most other modern audio compression schemes. He is commands. The main part of this paper reports acclaimed for pioneering work in digital audio coding, on a user survey on mappings between 22 DAW perceptual measurement techniques, Wave Field Syn - functions and 30 single-point as well as multi- thesis (WFS), and psychoacoustics. His honors include point gestures. We discuss the findings and the AES Silver Medal, the “IEEE Masaru Ibuka Con - point out consequences for user-interface sumer Electronic Award,” the German Future Award, design. which he shared with his colleagues, and the Cross of Convention Paper 8456 the Order of Merit of the Federal Republic of Germany. Furthermore he is a member in the “Hall of Fame” of the Consumer Electronics Association and of the Internation - Student/Career Development Event al Electrotechnical Commission. In 2009 he was appoint - RECORDING COMPETITION—PART 1 ed as Ambassador of the European Year of Creativity Saturday, May 14, 16:30 – 18:30 and Innovation. Brandenburg holds a Doctorate in Elec - Room 3 trical Engineering from Friedrich-Alexander University Erlangen-Nuremberg and received honorary Doctorate The Student Recording Competition is a highlight at each degrees from the universities Koblenz-Landau and convention. A distinguished panel of judges participates Lüneburg for his outstanding research work in the field of in critiquing finalists of each category in an interactive audio coding. He is professor at the Institute for Media presentation and discussion. This event presents stereo Technology at Ilmenau University of Technology and and surround recordings in these categories: director of the Fraunhofer Institute for Digital Media • Sound for Visual Media 16:30 to 17:30 Technology IDMT in Ilmenau, Germany. He is married to Ines Rein-Brandenburg. They share their home in Ilme - • Traditional Acoustic Recording 17:30 to 18:30 nau with two lovely cats. The title of his speech is “How to Provide High Quality Audio Everywhere: The MP3 The top three finalists in each category, as identified by Story and More …” our judges, present a short summary of their production Once upon a time there was the challenge to transmit intentions and the key recording and mix techniques high quality audio over phone lines. While this seemed used to realize their goals. They then play their projects impossible, ideas from psychoacoustics and signal pro - for all who attend. Meritorious awards are determined cessing—work by many researchers—helped the seem - here and will be presented at the closing Student Dele - ingly impossible to become reality: MP3 and other audio gate Assembly Meeting (SDA-2) on Monday afternoon. codecs enabled seamless transport of audio over thin The competition is a great chance to hear the work of lines. However, the MP3 story did not end there. The your fellow students at other educational institutions. Internet was changing shape, transforming from a text- Everyone learns from the judges’ comments even if your based medium into a major carrier for sound of all kinds, project isn’t one of the finalists, and it’s a great chance to including music. This meant changes not only for the meet other students and faculty members. payload (from text to audio), but also meant new dangers ¯

Audio Engineering Society 130th Convention Program, 2011 Spring 25 Technical Progra m for the audio quality delivered to music lovers, and it 09:30 changed business models for music sales dramatically, P14-2 Control of the Beamwidth of a Beamformer shaking up the music industry. with a Fixed Array Configuration— Today the challenges are different: we have a multi - Wan-Ho Cho ,1 Marinus M. Boone ,2 Jeong-Guon Ih ,3 tude of (legal) sources of music, and many of us have 1 access to Terabytes of music. How do we find our way Takeshi Toi 1Chuo University, Tokyo, Japan through this abundance of available content; how do we 2 find the gems in the millions of medium quality music? Delft University of Technology, Delft, The Netherlands Even then, audio reproduction still is far from perfect, so 3 how can we really fulfill the dream of complete auditory Korea Advanced Institute of Science and illusion, and what are the problems even today? The talk Technology (KAIST), Daejeon, Korea will introduce some current work on both MIR (Music The directional characteristic of the optimal Information Retrieval), on audio reproduction and the beamformer of a transducer array depends not psychoacoustic research needed to get further along to - only on its hardware configuration but also on ward the dream of perfectly reconstructed sound. the stability factor. This parameter can be used Brandenburg’s presentation will be followed by a to control the directivity of the array. In this paper reception hosted by the AES Technical Council. a method, which is based on the proper selec - tion strategy of the stability factor, is suggested to control the directional characteristics of the Student/Career Development Event optimal beamformer without changing the array STUDENT PARTY configuration. The selection method of the stabil - Saturday, May 14, 20:00 – 22:00 ity factor was investigated considering the trade- Join us for a fun and exciting evening at the Student off relation between spatial resolution and noise Social. On Saturday night, May 14, at 8 pm the fun amplification or array gain. The suggested begins! Don’t miss this chance to meet and engage with method was applied to the problems of both students from all over the world! Bring your own tracks microphone and loudspeaker arrays to obtain a and we will play them! Later, the local Student Section specific directivity pattern of high resolution with will guide us through the night. More information and a constant beamwidth. details will be provided at SDA-1. Convention Paper 8401

10:00

Session P14 Sunday, May 15 09:00 – 12:30 P14-3 Design 3-D High Order Ambisonics Encoding Room 1 Matrices Using Convex Optimization— Haohai Sun, U. Peter Svensson , Norwegian University of Science and Technology, Trondheim, Norway MULTICHANNEL AND SPATIAL SIGNAL In this paper we propose a convex optimization PROCESSING, PART 2 method for the design of 3-D High Order Ambisonics (3-D HOA) encoding matrices using Chair: Russell Mason , University of Surrey, Guildford, spherical microphone arrays, which offers the Surrey, UK possibility to impose spatial stop-bands in the directivity patterns of all the spherical harmonics while keeping the transformed audio channels still compatible with the 3-D HOA reproduction 09:00 sound format. Using the proposed convex opti - P14-1 Spatial Analysis of Room Impulse Responses mization method, the globally optimal encoding Captured with a 32-Capsule Microphone matrices can be obtained, and the suitable Array— Angelo Farina, Alberto Amendola, trade-off among several design factors, e.g., Andrea Capra, Christian Varani , University of response distortions in the obtained spherical Parma, Parma, Italy harmonic, the dynamic range of matrix coeffi - cients, i.e., the system robustness, and frequen - The authors developed a new measurement cies, can be analyzed and illustrated. The pro - system, which captures 32-channel impulse posed convex optimization is formulated as a responses by means of a spherical microphone form of second-order cone programming that array and a matrix of FIR filters, capable of prov - can be efficiently solved. Numerical results vali - ing frequency-independent directivity patterns. date the proposed method. This method can be This allows for spatial analysis with resolution easily generalized to the 2-D HOA cases. much higher than what was possible with obso - Convention Paper 8402 lete sum-and-delay beamforming. The software developed for this application creates a false- 10:30 color video of the spatial distribution of energy, changing with running time along the impulse P14-4 Principles in Surround Recordings with response duration. A virtual microphone probe Height— Günther Theile ,1 Helmut Wittek 2 allows extraction of the sound coming from any 1VDT, Geretsried, Germany specific direction. The method was successfully 2Schoeps Mikrofone, Karlsruhe, Germany employed in three concert halls, providing guid - ance for correcting some acoustical problems New multichannel sound formats extending 5.1 (echo, focusing) and for placing sound reinforce - with height channels are adding the third dimen - ment loudspeakers in optimal positions. sion to recordings. They provide a much wider Convention Paper 8400 range of spatial sound effects and allow more real - ism of spatial reproduction in terms of direct

26 Audio Engineering Society 130th Convention Program, 2011 Spring sound, early and late reflections, reverberation, University of Science & Technology, Clearwater and ambience sound. Using the example of two Bay, Kowloon, Hong Kong upper layer front and two upper layer surround complementary loudspeakers (5.1+4, known as The “delay-and-add” theory (Hebrank and “Auro 3D 9.1”) the psychoacoustic principles in the Wright, 1974) was used to calculate 4068 perception of elevated phantom sound sources, matching scores between each of the 192 spatial depth, spatial impression, envelopment, HRTFs and the dimensions of 12 pairs of ears ambient atmosphere, as well as directional stability for two incident sound directions (30 degrees up within the sweet area are discussed. Concrete and 30 degrees down). Five HRTFs with 0, 25, proposals for microphone configurations can 50, 75, and 100th percentile average matching evolve from these considerations. scores were selected for two incident directions. Convention Paper 8403 These 10 HRTFs were used to produce 10 sound cues (5 from 30 degrees up and 5 from 30 degrees down). Ten listeners participated in a 11:00 sound localization experiment to localize the 10 P14-5 Efficient 3-D Sound Field Reproduction— sound cues presented in random order and with Mincheol Shin, 1 Filippo Fazi ,1 Jeongil Seo ,2 5 repetitions. Preliminary results indicated that Philip A. Nelson 1 matching scores can explain up to 22% of the 1ISVR, University of Southampton, inter-subject variations in localization errors. Southhampton, UK Potential applications are discussed. 2Electronics Telecommunication Research Convention Paper 8406 Institute, Daejeon, Korea A method is presented for efficient sound field Workshop 9 Sunday, May 15 reproduction with a loudspeaker array constitut - 09:00 – 11:00 Room 2 ed by multiple sound sources that are three- dimensionally distributed. The physical reproduction of several target sound fields is in - WHAT EVERY SOUND ENGINEER SHOULD KNOW vestigated when the target sound fields are sur - ABOUT THE VOICE rounded by multiple sound sources. A cost func - tion to be minimized has been developed to Chair: Eddy B. Brixen , EBB Consult, Denmark, obtain the optimal solution with reasonable ener - TC-MA and TC-AF gy distribution and sufficient sweet area when the distribution of loudspeakers is non-uniform. The performance of the proposed method is ver - Panelists: Evelyn Abberton , Department of Phonetics ified by the results of computer simulations and and Linguistics University College, London, UK subjective tests in the cases of NHK 22.2 and Adrian Fourcin , Department of Phonetics ETRI 10.2 channel configurations. and Linguistics University College, London, UK Convention Paper 8404 Henrik Kjelin , Vocal Institute, Denmark Julian McGlashan , Department of Otorhinolaryngology, Queen’s Medical Centre 11:30 Campus, Nottingham University Hospitals, UK P14-6 On the Scattering of Synthetic Sound Mikkel Nymand , Timbre Music / DPA Fields— Jens Ahrens, Sascha Spors , Deutsche Microphones, Denmark Telekom Laboratories, Technische Universität Cathrine Sadolin , Complete Vocal Institute, Berlin, Berlin, Germany Denmark In sound field synthesis a given arrangement of The purpose of this workshop is to teach sound engi - elementary sound sources is employed in order neers how to listen to the voice before they even think of to synthesize a sound field with given desired microphone picking and knob-turning. physical properties over an extended region. The presentation and demonstrations are based on The calculation of the driving signals of these the “Complete Vocal Technique” (CVT) where the funda - secondary sources typically assumes free-field mental is the classification of all human voice sounds propagation conditions. The present paper in - into one of four vocal modes named Neutral, Curbing, vestigates the scattering of such synthetic sound Overdrive, and Edge. The classification is used by pro - fields from unavoidable scattering objects like fessional singers within all musical styles, and has in a the head and body of a person apparent in the period of 20 years proved easy to grasp in both real life target region. It is shown that the basic mecha - situations and also in auditive and visual tests (sound nisms are similar to the scattering of natural examples and laryngeal images/Laryngograph ® wave - sound fields. Though, synthetic sound fields can forms). These vocal modes are found in the speaking exhibit properties different to those of natural voice as well. sound fields. Consequently, in such cases also Cathrine Sadolin, the developer of CVT, will involve the scattered synthetic sound fields exhibit prop - the audience in this workshop, while explaining and erties different to those of scattered natural demonstrating how to work with the modes in practice to sound fields. achieve any sound and solve many different voice prob - Convention Paper 8405 lems like unintentional vocal breaks, too much or too little volume, hoarseness, and much more. Julian McGlashan, 12:00 Adrian Fourcin, and Evelyn Abberton will explain the physical aspects of the voice and demonstrate laryngo - P14-7 Toward Mass-Customizing Up/Down Generic graph waveforms and analyses. Mikkel Nymand will 3-D Sounds for Listeners: A Pilot Experiment explain essential parameters in the recording chain— Concerning Inter-Subject Variability— John especially the microphone—to ensure reliable and natur - Au, Richard So, Andrew Horner , The Hong Kong al recordings. ¯ Audio Engineering Society 130th Convention Program, 2011 Spring 27 Technical Progra m Tutorial 5 Sunday, May 15 – Source directivity, positioning and aiming 09:00 – 10:30 Room 3 – Discrete reflections and how to avoid them AUDIO OVER IP—THE BASICS – What to include in a new venue for optimum adaptation for reinforced sound Peter Stevens Presenter: , BBC R&D, London, UK – Practical measures in existing situations With the slow demise of ISDN connections and technolo - – Workarounds in acoustic harsh environments. gy, audio over IP plays an increasingly important role for transporting audio signals from A to B. But this is only Tutorial 7 Sunday, May 15 one area where audio over IP is used. Many more poten - 11:00 – 12:30 Room 3 tial and also existing applications show where we are heading. This tutorial looks at the basics of this technolo - gy and pinpoints the advantages as well as the pitfalls LOUDNESS AND EBU R 128—A BASIC TUTORIAL and how they can be avoided. Practical examples will OF THE WHY AND WHAT underline the concepts. Presenter: Florian Camerer , ORF, Austrian TV, Tutorial 6 Sunday, May 15 and Chairman of EBU PLOUD 09:00 – 10:30 Room 4 The EBU recommendation R 128 “Loudness normaliza - tion and permitted maximum level of audio signals” is a NO—I SAID “INTERACTIVE” MUSIC milestone for establishing a new way to level audio sig - nals. Together with four supporting documents it pro - Co-Chairs: Dave Raybould , Leeds Metropolitan vides the framework for the transition into this new world University, Leeds, UK of leveling audio based on loudness, not peaks. The Richard Stevens , Leeds Metropolitan chairman of the EBU group PLOUD will present R 128 in University, Leeds, UK detail with the help of examples and will give an outlook of how and where the recommendation is already in This session will recap a number of common approaches practical use. to so called “interactive” music in games through a series of practical in-game demonstrations. We’ll discuss what Sunday, May 15 11:00 Room Saint Julien is understood by the terms reactive, adaptive, and inter - Technical Committee Meeting on Hearing active and put forward the argument that there's actually and Hearing Loss Prevention very little about the current use of music in games that is truly “interactive.” The session will conclude with further Sunday, May 15 11:00 Room Mouton Cadet examples of some potential future solutions of how we Standards Committee Meeting SC-02-01 on Digital might better align the time-based medium of music with Audio Measurement Techniques the nonlinear medium of games. This session is intended to be accessible to complete beginners but also thought Special Event provoking to old pros! AES/APRS/DV247 EVENT: TALKBACK-PRO 2 LEAD ME DOWN THE SIGNAL PATH Sunday, May 15 09:00 Room Mouton Cadet Sunday, May 15, 11:15 – 13:00 Standards Committee Meeting SC-04-04 on Room 2 Microphone Measurement and Characterization Moderator: Gil Limor Sunday, May 15 10:00 Room Saint Julien Panelists: TBA Technical Committee Meeting on Loudspeakers and Headphones “All you need is a great mic and a decent pre-amp and you can expect your home recordings to match any stu - Workshop 10 Sunday, May 15 dio.” A familiar claim but a pretty glib observation, espe - 11:00 – 12:30 Room 5 cially when there are so many “affordable” options and such a wide diversity of “subjective” opinions. “Listen, position, and condition” is the mantra of an engi - ACOUSTICS FOR SOUND REINFORCEMENT neer trying to capture a live sound accurately. How does each skill impact on the others? Can the flaws in a cheap Chair: Peter Mapp , Peter Mapp Associates, microphone be rectified by cunning signal manipulation or Colchester, Essex, UK does the notion of “less is more” apply to signal path? A group of key experts from academics to manufactur - Panelist: Ben Kok , Ben Kok – Acoustic Consulting ers discuss the fundamentals of audio capture and re - finement. Traditionally, acoustic design for performance spaces focuses on optimum acoustics for non-reinforced perfor - Session P15 Sunday, May 15 11:30 – 13:00 mances. However, a fine concert hall for symphonic Foyer music can be very troublesome for reinforced sound. Even performance spaces designed with variable acoustics are mostly optimized for acoustic sources, little POSTERS: SPEECH AND CODING consideration is given to the needs for reinforced sound. This session addresses the different acoustic require - 11:30 ments for reinforced sound as opposed to non-reinforced sound. Topics to be discussed will include: P15-1 A New Robust Hybrid Acoustic Echo Cancellation Algorithm for Speech – Reverberation, support or interference?

28 Audio Engineering Society 130th Convention Program, 2011 Spring Communication in Car Environments— Yi partially contradictory to previous studies. This Zhou, Chengshi Zheng, Xiaodong Li , Chinese leads to the conclusion that perceived quality is Academy of Sciences, Beijing, China strongly linked to the focus, background, and abilities of the assessors. The test design, real - This paper studies a new robust hybrid adaptive ization, and obtained results are shown, as well filtering algorithm for acoustic echo cancellation as a comparison to studies conducted with dif - (AEC) in a car speech communication system. ferent types of users. The proposed algorithm integrates the affine Convention Paper 8409 projection algorithm (APA) and normalized least mean square (NLMS) algorithm. It can switch 11:30 between them with a coefficient derivative-based technique to achieve overall optimum conver - P15-4 Analysis of Parameters of the Speech Signal gence performance. To help the algorithm com - Loudness of Professional Television bat deteriorating impulsive interferences widely Announcers— Mirjana Mihajlovic ,1 Dejan Todor - encountered in car environments and enhance ovic ,2 Iva Salom 3 the system robustness in double talk period, 1Radio Television of Serbia, Belgrade, Serbia robust statistics technique is also incorporated 2Radio Belgrade, Belgrade, Serbia into the proposed algorithm. Experiments are 3Institute Mihajlo Pupin, Belgrade, Serbia conducted to verify the improved and robust overall AEC convergence performance achieved The paper analyzes speech loudness of profes - by the new algorithm. sional announcers, who do large percentage of Convention Paper 8407 television audio signals. Over 100 files of speech signals recorded with same equipment were 11:30 analyzed. For each file the authors calculated loudness and true peak, according to ITU-R BS P15-2 Speech Source Separation Using a Multi- 1770 and EBU R128 and RMS. There were cri - Pitch Harmonic Product Spectrum-Based teria adopted for classification of records, and Algorithm — Rehan Ahmed, Roberto Gil-Pita, results were statistically analyzed. Analysis David Ayllón, Lorena Álvarez , University of showed that the loudness of the speech signal Alcalá, Alcalá de Henares, Spain depends on the gender of speakers, intonation, and type of program material, and also that files This paper presents an efficient algorithm for of greater loudness may not always have higher separating speech signals by determining multi - peak value to files of lower loudness. It was not - ple pitches from mixtures of signals and assign - ed there was a significant overlap among the ing the sources to one of those estimated short-term loudness and RMS values calculated pitches. The pitch detection algorithm is based with the same time constant. on Harmonic Product Spectrum. Since the pitch Convention Paper 8410 of speech signals fluctuates readily, a frame- based algorithm is used to extract the multiple 11:30 pitches in each frame. Then, the fundamental frequency (pitch) for each source is estimated P15-5 Improved Prediction of Nonstationary and tracked after comparing all the frames. The Frames for Lossless Audio Compression— estimated fundamental frequency of the sources Florin Ghido , Tampere University of Technology, is then used to generate a set of binary masks Tampere, Finland that allow separating the signals in the Short Time Fourier Transform domain. Results show a We present a new algorithm for improved predic - considerable separation of the speech signals, tion of nonstationary frames for asymmetrical justifying the feasibility of the proposed method. lossless audio compression. Linear prediction is Convention Paper 8408 very efficient for decorrelation of audio samples, however it requires segmentation of the audio 11:30 into quasi-stationary frames. Adaptive segmen - tation tries to minimize the total compressed P15-3 User-Orientated Subjective Quality size, including the quantized prediction coeffi - Assessment in the Case of Simultaneous cients for each frame, thus longer frames that Interpreters— Judith Liebetrau, 1 Thomas are not quite stationary may be selected. The Sporer ,1 Paolo Tosoratti ,2 Daniel Fröhlich, 1 new algorithm for computing the linear prediction Sebastian Schneider, 1 Jens-Oliver Fischer 1 coefficients improves compressibility of nonsta - 1Fraunhofer Institute for Digital Media tionary frames when compared with the least Technology IDMT, Ilmenau, Germany squares method. With adaptive segmentation, 2DG Interpretation, Brussels, Belgium the proposed algorithm leads to small but con - sistent compression improvements up to 0.56%, A study was conducted to explore various sub - on average 0.11%. For faster encoding using jective quality aspects of audio-visual systems fixed size frames, without including adaptive seg - for interpreters. The study was designed to bring mentation, it significantly reduces the penalty on the existing requirements for on-site simultane - compression with more than 0.21% on average. ous interpretation up-to-date and to specify new Convention Paper 8411 requirements for remote interpretation (telecon - ferencing). The feasibility of using objective 11:30 measurement methods in this context was examined. Several parameters influencing P15-6 A Lossless/Near-Lossless Audio Codec for perceived quality, such as audio coding, video Low Latency Streaming Applications on coding, room characteristics, and audio-visual Embedded Devices— Neil Smyth , Cambridge latency were assessed. The results obtained are ¯

Audio Engineering Society 130th Convention Program, 2011 Spring 29 Technical Progra m Silicon Radio, Belfast, Northern Ireland Sunday, May 15 13:00 Room Saint Julien Technical Committee Meeting on High Resolution Increasingly there is a need for high quality Audio audio streaming in devices such as modular home audio networking systems, PMPs, and Sunday, May 15 13:00 Room Mouton Cadet wireless loudspeakers. As wireless transmission Standards Committee Meeting SC-04-03 on capacities continue to increase it is desirable to Loudspeaker Modeling and Measurement utilize this increasing data bandwidth to perform real-time wireless streaming of audio content coded in a lossless or near-lossless format. In Session P16 Sunday, May 15 14:00 – 17:30 this paper we discuss the development of an Room 1 adaptive audio coding algorithm that balances the design goals of low latency, low complexity, error robustness, and a dynamically-variable bit AUDIO SIGNAL PROCESSING AND ANALYSIS rate that scales to mathematically-lossless cod - ing under suitable conditions. Particular empha - Chair: Jayant Datta sis is placed on the optimization of the algorithm structure for real-time audio processing applica - 14:00 tions and the mechanism by which “hybrid” loss - less and near-lossless coding is achieved. P16-1 A New Approach to Designing Decimation Convention Paper 8412 Filters for Oversampled A/D Converters— Jamie A. S. Angus , University of Salford, 11:30 Salford, Greater Manchester, UK P15-7 Low-Delay Directional Audio Coding for Real- This paper presents a new approach to design - Time Human-Computer Interaction— Tapani ing the necessary decimation filter in over-sam - Pihlajamäki, Ville Pulkki , Aalto University School pled noise shaping analog to digital converters. of Electrical Engineering, Espoo, Finland These filters are still finite impulse response designs but they are designed using novel win - Games and virtual worlds require low-cost, dow functions that produce a stop-band attenua - good-quality, and low-delay audio engines. The tion that increases with frequency thus matching Short-Time Fourier Transform-based (STFT) the out of band noise characteristics of the implementation of Directional Audio Coding noise-shaping modulator. As a result high quality (DirAC) for virtual worlds fulfills the two first decimation filters can be realized using shorter demands. Unfortunately, the delay can be per - filter lengths. ceivably large. In this paper, a modification to Convention Paper 8414 DirAC is introduced that uses different time-fre - quency resolutions for STFT concurrently, which 14:30 are not aligned in time in reproduction. This leads to the high-frequency non-diffuse content P16-2 Warped IIR Filter Design with Custom being reproduced as soon as possible and thus Warping Profiles and its Application to Room reduces the perceived delay. Informal tests Response Modeling and Equalization— Balázs show that the delay was short enough for a Bank , Budapest University of Technology and musician to play an instrument through the pro - Economics, Budapest, Hungary cessing. Moreover, the results of formal listening tests show that the reduction in quality is percep - In traditional warped FIR and IIR filters, the fre - tible but not annoying. quency-warping profile is adjusted by a single Convention Paper 8413 free parameter, leading to a less flexible alloca - tion of frequency resolution. As an example, it is not possible to achieve a truly logarithmic fre - quency resolution, which would be often desired Workshop 11 Sunday, May 15 in audio applications. In this paper a new 11:30 – 13:00 Room 4 approach is presented for warped IIR filter design where the filter specification is trans - EMERGING TRENDS IN AUDIO FOR GAMES formed by any desired (e.g., logarithmic) fre - quency transformation, and a standard IIR filter is designed to this transformed specification. Presenters: Michael Kelly , Sony Computer Then, the poles and zeros of this transformed fil - Entertainment Europe, London, UK ter are found and mapped back to the original Steve Martz , THX, CA, USA frequency scale. Due to the approximations in mapping back the poles and zeros, the resulting This workshop looks at the current state of technology transfer function may show some discrepancies requirements for audio in game applications and discuss - from its optimal version. This is resolved by an es emerging trends and the technical requirements im - additional optimization of the zeros of the final fil - posed by those trends. The event is presented by the ter. Examples of loudspeaker-room response Co-Chairs of the AES Technical Committee on Audio for modeling and equalization are presented. Games. The workshop also summarizes some of the Convention Paper 8415 topics presented at the recent 41st International Confer - ence on Audio for Games.. 15:00 P16-3 Computationally Efficient Nonlinear Sunday, May 15 12:00 Room Saint Julien Chebyshev Models Using Common-Pole Technical Committee Meeting on Signal Processing Parallel Filters with the Application to

30 Audio Engineering Society 130th Convention Program, 2011 Spring Loudspeaker Modeling— Balázs Bank , ean distance in MFCC feature space. This met - Budapest University of Technology and ric is justified on the basis of its success in previ - Economics, Budapest, Hungary ous work. The genetic algorithm offers the best results with the FM and subtractive test synthe - Many audio systems show some form of nonlin - sizers but the hill climber and data driven ear behavior that has to be taken into account in approach also offer strong performance. The modeling. For this, often a black-box model is concept of sound synthesis error surfaces, identified, coming from the generality and sim - allowing the detailed description of sound syn - plicity of this approach. One such model is the thesis space, is introduced. The error surface for simplified Volterra model, using parallel branch - an FM synthesizer is described and suggestions es that have a polynomial-type nonlinearity and are made as to the resolution required to effec - a linear filter in series. For example, Chebyshev tively represent these surfaces. This information models use Chebyshev polynomials as nonlin - is used to inform future plans for algorithm ear functions, making the model identification a improvements. very straightforward procedure by using logarith - Convention Paper 8418 mic sweep measurements. This paper proposes a highly efficient implementation of Chebyshev 16:30 models by using fixed-pole parallel filters for the linear filtering part. The efficiency comes from P16-6 On the Multichannel Sinusoidal Model for the fact that parallel filters can have a logarith - Coding Audio Object Signals— Toni mic frequency resolution, which better fits the Hirvonen ,1 Athanasios Mouchtaris 1,2 human hearing behavior than traditional FIR or 1Institute of Computer Science, Foundation for IIR filters. Moreover, the branches can share the Research and Technology–Hellas same denominators, leading to an additional (FORTH-ICS) Heraklion, Crete, Greece performance benefit. The proposed model is (now with Dolby Laboratories, Stockholm, particularly well suited for the real-time digital Sweden) simulation of loudspeakers and other weakly 2University of Crete, Heraklion, Crete, Greece nonlinear devices, such as tube guitar amplifiers. Convention Paper 8416 This paper presents two improvements on a recently proposed multichannel sinusoidal mod - 15:30 eling system for coding multiple audio object signals. The system includes extracting the P16-4 Event-Driven Real-Time Audio Processing sinusoidal components and an LPC envelope with GPGPUs— Tiziano Leidi, 1 Thierry Heeb, 1 for each object signal, as well as transform cod - Marco Colla ,1 Jean-Philippe Thiran 2 ing of the residuals’ downmix. The contributions 1ICIMSI-SUPSI, Manno, Switzerland of this paper are: (a) a psychoacoustic model for 2EPFL, Lausanne, Switzerland enabling the system to scale well with multiple object signals, and (b) an improved method to Development of real-time audio processing encode the common residual, tailored to the applications for GPGPUs is not without chal - “white” nature of this signal. As a result, sound lenges. Parallel processing of audio signals is quality of around 90% on the MUSHRA scale is often constrained by serial dependencies within obtained for 10 simultaneous object signals cod - or between the algorithms. On GPGPUs, insuffi - ed with a total rate of 150 kbit/s, while retaining cient data pressure further limits the attainable the individual object parametric representations. performance improvements, as it causes inactiv - Convention Paper 8419 ity of the GPU cores. In this paper we analyze the limits of audio processing on GPGPUs and 17:00 present an approach based on event-driven scheduling, that maximizes data pressure to P16-7 An Additive Synthesis Technique for favor performance improvements. We also pre - Independent Modification of the Auditory sent recent enhancements of Audio n-Genie, an Perceptions of Brightness and Warmth— open-source development environment for Asteris Zacharakis, Joshua Reiss , Queen Mary audio-processing applications. By combining University of London, London, UK Audio n-Genie and the proposed approach, we show that it is possible to increase audio pro - An algorithm that achieves independent modifi - cessing speed-up. cation of two low-level features that are correlat - Convention Paper 8417 ed with the auditory perceptions of brightness and warmth was implemented. The perceptual 16:00 validity of the algorithm was tested through a series of listening tests in order to examine P16-5 A Comparison of Parametric Optimization whether the low-level modification was indeed Techniques for Tone Matching— Matthew Yee- perceived as independent and to investigate the King ,1 Martin Roth 2 influence of the fundamental frequency on the 1Goldsmiths, University of London, London, UK perceived modification. A Multidimensional Scal - 2Reality Jockey Ltd., London, UK ing analysis (MDS) on listener responses to pair - wise dissimilarity comparisons accompanied by Parametric optimization techniques are com - a verbal elicitation experiment examined the per - pared in their abilities to elicit parameter settings ceptual significance and independence of the for sound synthesis algorithms, which cause two low-level features chosen. This is a first step them to emit sounds as similar as possible to for the future development of a perceptually target sounds. A hill climber, a genetic algorithm, based control of an additive synthesizer. a neural net, and a data driven approach are Convention Paper 8420 compared. The error metric used is the Euclid - ¯

Audio Engineering Society 130th Convention Program, 2011 Spring 31 Technical Progra m Session P17 Sunday, May 15 14:00 – 15:30 was analyzed using histograms and statistical Foyer quantities (moments), such as the kurtosis and skewness. In this paper the above analysis is POSTERS: BINAURAL AND SPATIAL AUDIO extended to binaural room impulse responses (BRIRs) and the dependence of the statistical measures on the room acoustical properties, 14:00 such as the reverberation time, the room size, and the source-receiver distance is discussed. P17-1 Repeatability of Localization Cues in HRTF Emphasis is given on the binaural measure of Data Bases— Elena Blanco-Martin, Silvia the magnitude squared coherence (MSC), which Merino is considered to be an important cue for binaural Saez-Miera, Juan José Gomez-Alfageme, Luis hearing related to perceptual aspects such as Ignacio Ortiz-Berenguer , Universidad Politecnica the source width, the envelopment and the spa - de Madrid, Madrid, Spain ciousness. After a brief overview of the existing Head-Related Transfer Function (HRTF) repre - MSC models, a perceptually-motivated MSC sents the time-spectral filtering that head and implementation is examined, based on a gam - torso do to a sound that goes from a sound matone filterbank. MSC results for various source to the ears. These transfer functions rooms and source-receiver positions are pre - bring the localization cues that vary according to sented and related to the existing MSC models. sound source position (azimuth and elevation). Convention Paper 8423 Human auditory system uses such localization cues for estimating the sound direction. HRTF 14:00 are used in two ways. One way is for synthesiz - P17-4 Spatial Sound and Stereoscopic Vision— ing binaural sound (virtual audio 3-D). The sec - Paul ond way is for analyzing binaural sounds in Mannerheim , University of California, Santa Bar - order to estimate the localization of sound bara, Santa Barbara, CA, USA sources. Therefore HRTF are important and This paper presents a technique for reproducing useful data for researchers. There are few public coherent audio visual images for multiple users, HRTF data bases. The most known is the Kemar only wearing 3-D glasses and without utilizing Data Base from MIT. There is also the Cipic head racking. The recent emergence of 3-D con - Data Base from UC Davis and the Itakura Data tent has increased the demand for technology Base. In this paper we present a new public data that can display visual images that are coherent base for researching use available on the web. with sound images for multiple users. Audio The data base has been measured on the Head visual object difference is here investigated for and Torso Simulator 4100 by Brüel & Kjær. In analyzing the size of the sweet spot of a system addition, a comparative study of localization that combines a visual display technique named cues from the data bases is carried out for show - stereoscopy with a sound reproduction tech - ing their repeatability. nique called wave field synthesis. The sweet Convention Paper 8421 spot of such a configuration is limited due to dif - ferences in characteristics between the sound 14:00 reproduction system and the visual display; how - ever as a consequence, it is found that the num - P17-2 Dynamic Head-Related Transfer Function ber sources in the wave field synthesis array can Measurement Using a Dual-Loudspeaker be reduced. Array— Qinghua Ye, Qiujie Dong, Lingling Convention Paper 8424 Zhang, Xiaodong Li , Institute of Acoustics, Chinese Academy of Sciences, Beijing, China 14:00 A dynamic Head-Related Transfer Function P17-5 Designing Ambisonic Decoders for Improved (HRTF) measurement method using a dual-loud - Surround Sound Playback in Constrained speaker array is presented, which reduces mea - Listening Spaces— 1 suring requirements and increases the efficiency David Moore , Jonathan Wakefield 2 as well. First, the dual-loudspeaker array emits 1 uncorrelated signals, while head size and head Glasgow Caledonian University, Glasgow, Scotland, UK motion are obtained through short-time time- 2 delay estimation. Second, in the approximation University of Huddersfield, Huddersfield, UK of linear time-invariant (LTI) within the testing Much research has been undertaken to optimize time, multi-point continuous HRTF measurement irregular 5-speaker Ambisonic decoders for ide - is accomplished. Compared with FASTRAK alized listening environments. In such environ - head tracking system, the experimental results ments loudspeaker placement is not restricted confirm the validity of the proposed method. and can conform to the ITU 5.1 standard. In do - Convention Paper 8422 mestic settings, the room shape, furniture, and television positioning often restrict speaker 14:00 placement. It is often the case that a compro - mised speaker layout is enforced by other P17-3 Statistical Analysis of Binaural Room domestic requirements. This paper seeks to de - Impulse Responses— Eleftheria Georganti, rive Ambisonic decoders to optimize perceived Alexandros Tsilfidis, John Mourjopoulos , localization performance for these constrained University of Patras, Patras, Greece asymmetrical speaker layouts. This work uses a In previous work of the authors, the spectral heuristic search algorithm to derive decoder magnitude of room transfer functions (RTFs) coefficients and simultaneously optimize loud -

32 Audio Engineering Society 130th Convention Program, 2011 Spring speaker angle within specified bounds. Theoreti - sound field reproduction capabilities of different cal results are shown for different orders of loudspeaker setups. We present a model, based newly derived Ambisonic decoders for typical on interaural coherence and interaural level dif - domestic scenarios. ference, for estimating perceived diffuseness of Convention Paper 8425 synthetic sound fields evoked by an arbitrary number of transducers at different positions. The 14:00 results of different loudspeaker setups and lis - tener orientations are compared. P17-6 Decoding for 3-D— Johannes Boehm , Convention Paper 8428 Technicolor, Research & Innovation, Hannover Germany Three dimensional spatial sound reproduction Workshop 12 Sunday, May 15 using irregular loudspeaker layouts requires a 14:00 – 16:00 Room 2 special decoder design. We present the funda - mentals of Higher Order Ambisonics (HOA) DEAF AID LOOPS AND ASSISTIVE LISTENING decoding. Then we focus on beam forming tech - SYSTEMS—UNDERSTANDING THE NEEDS niques to derive solutions for irregular spaced AND THE TECHNOLOGY setups. Panning functions are created by vector base amplitude panning or least-squares meth - Chair: Peter Mapp , Peter Mapp and Associates, ods as patterns and are modeled by spherical Colchester, UK harmonics. The required HOA order for effective beam forming proves to be in inverse proportion to the minimal angular spacing of speakers. We Panelists: Conny Anderson , Univox demonstrate decoder design using an example Doug Edworthy , DGE Associates setup of 16 speakers, and evaluate objective Ken Hollands , Ampetronic performance criteria. We conclude that decoders John Woodgate , J M Woodgate & Associates for irregular setups require higher HOA orders Although deaf aid loop systems have been around for a compared to decoders for regular setups using considerable period of time, they are still not well under - the same number of speakers and discuss the stood, and many systems do not achieve an acceptable consequences. standard. Deaf aid loops—or AFILS as they are formally Convention Paper 8426 known—are installed in all forms of venue or public build - ing raging from churches, theaters, and cinemas to rail - 14:00 way and underground stations. Loop systems can vary from large systems covering many hundreds of seats to P17-7 Real-Time Reproduction of Moving Sound counter-top systems intended for individual users. The Sources by Wave Field Synthesis: Objective workshop will review the theory and practice of Audio and Subjective Quality Evaluation— Michele Frequency Induction Loop Systems (AFILS) as well as Gasparini, Paolo Peretti, Stefania Cecchi, Laura discussing other technologies such as infra red. Ways of Romoli, Francesco Piazza , Universita optimizing system performance and day-to-day problems Politecnica delle Marche, Ancona, Italy will be discussed. Wave Field Synthesis (WFS) is an audio render - ing technique that allows the reproduction of an Tutorial 8 Sunday, May 15 acoustic image over an extended listening area. 14:00 – 15:30 Room 4 In order to achieve a realistic sensation, true representation of moving sound sources is es - VIDEO FOR AUDIO ENGINEERS sential. In this paper a real time algorithm that significantly reduces computational efforts in the Presenter: David Tyas , Ikon, Worcester, UK synthesis of WFS driving functions in presence of moving sound sources is presented. High effi - Incorporating video into a system can be a good addi - ciency can be obtained taking into account a tional source of revenue for audio professionals. It may model based on phase approximation and Short be simply adding a projector to a theatre or digital Time Fourier Transform (STFT). The influence of signage to augment a VA system. But, with multiple ana - the streaming frame size and of the source logue video formats and a transition to an equally con - velocity on well known sound field artifacts has fusing array of incompatible digital formats, what con - been studied, considering PC simulations and nects together and works, how to you convert between listening tests. the formats and how do you cable and connect. Convention Paper 8427 This short tutorial will cover the basics of video with the emphasis on the newer formats, how to interface these 14:00 and how to incorporate legacy products into systems.

P17-8 Assessing Diffuse Sound Field Reproduction Sunday, May 15 14:00 Room Saint Julien Capabilities of Multichannel Playback Technical Committee Meeting on Audio for Games Systems— Andreas Walther, Christof Faller , Ecole Polytechnique Fédérale de Lausanne, Special Event Lausanne, Switzerland AES/APRS EVENT: The generation of subjectively diffuse sound THE RECORDING LEGACY OF PHIL SPECTOR fields is an essential part of creating pleasing Sunday, May 15, 14:30 – 16:15 synthetic sound fields using loudspeaker play - Room 3 back. A number of studies have been published Moderator: Barry Marshall presenting subjective evaluations of the diffuse ¯

Audio Engineering Society 130th Convention Program, 2011 Spring 33 Technical Progra m The First Producer? Phil Spector and the Legacy An area of digital audio manipulation currently in of the Svengali Producer flux is that of time stretching. Following the emergence of real-time granular synthesis as a This lecture will explore the career of Phil Spector, the compositional tool, early sampler-based imple - legendary figure who arguably created the role of the mentations were pushed beyond “authenticity” to record producer as we know it today. It will trace Spec - create new timbres in the commercial music of tor’s historical development from his early career as an the 1990s. As the algorithms improve, allowing artist, his apprenticeship with Atlantic Records and more flexible and transparent implementation Leiber & Stoller (the first credited producers), his flower - today, even more opportunities for a new “cre - ing as a producer-mogul with his Philles record label, his ative abuse” exist. This brief will first contextual - initial “retirement” in 1967, his re-invention as the produc - ize through consideration of the metaphor of er of John Lennon and George Harrison to his later spo - authenticity in the tape recording of the 1940s radic (yet sometimes still seminal) work with artists like and its soon-parallel abuse, which offered new the Ramones. The emphasis of the lecture will be on the pathways into multi-tracking and Musique Con - development of the role of the producer—through Spec - crète. The brief will chronologize, and continue tor’s initial creation of his “Wall of Sound” and his Sven - by examining potential for exploitation of stretch - gali approach to artist development—and then through ing artifacts in some contemporary algorithms, his sublimation of the Svengali role into a more mea - and propose a quantification of this effect. sured approach with his work with the gigantic talents of Engineering Brief 9 artists like Lennon and Harrison. The lecture will include lots of examples of Spector’s work. 16:00 Sunday, May 15 15:00 Room Saint Julien EB2-2 Activity Flow in Music Equalization: The Technical Committee Meeting on Acoustics Cognitive and Creative Implications and Sound Reinforcement of Interface Design— Joshua Mycroft, 1 Justin Paterson 2 Sunday, May 15 15:00 Room Mouton Cadet 1Queen Mary University of London, London, UK Standards Committee Meeting SC-03-06 and SC-03- 2University of West London, London, UK 07 Digital Library and Archive Systems and Audio The mixing desk metaphor found in many Digital Metadata (combined meeting) Audio Workstations (DAW) creates a quantitative visual display that is highly structured and seg - Workshop 13 Sunday, May 15 mented. While this is useful for transmitting large 15:45 – 16:45 Room 4 amounts of quantitative data it can inhibit the more intuitive and performative aspects inherent in IMPLEMENTATION OF INTERACTIVE MUSIC music mixing. This paper’s focus is the cognitive IN GAMES and creative issues encountered using current music production software equalizers and the Chair: Erasmus Talbot , Black Rock Studio, influence they exert on the initial approach, task Brighton, UK workflow, and final output of the user. Equalizers have been chosen to exemplify this, due to their This workshop looks at the implementation of interactive pivotal balance between aural and visual modali - music in current video games. It focuses on the ties. The paper draws conclusions as to the effec - approach taken in “Split/Second,” an arcade racing game tiveness of current software equalizer designs and packed with action of an epic scale. To reflect its allu - proposes modifications to design. sions to Hollywood blockbusters in audio, the team Engineering Brief 10 turned their backs to licensed music tracks and opted for a bespoke score. Initially all music was delivered exter - 16:00 nally, but the talk explains how the team became increasingly involved in the music production process. EB2-3 The Preset Is Dead; Long Live the Preset— This approach allowed the team to realize creative possi - Justin Paterson , University of West London, bilities that could only emerge from strong familiarity with London, UK the game. The workshop gives a detailed explanation of the nov - The use of preset sounds in audio production el production method and playback system used in has long been scorned by professional produc - Split/Second and highlights how bold choices, informed ers, some of whom have cited a lack of original - by a particular outlook on interactive music, can lead to a ity or integrity, or perhaps a proliferation of simple, yet effective smoke-and-mirror type solution. It homogenization in productions. Despite this, will also highlight some pitfalls that were encountered, manufacturers have continued to develop ever- which elicited future improvements to the system. larger ranges of presets, now extending beyond instruments and effects, to EQ and even whole channel strips. Developmental work continues Session EB2 Sunday, May 15 16:00 – 17:30 to further automate aspects of the mixing Foyer process itself. This brief will examine the impli - cations of presets from yesterday to today, and ENGINEERING TOOLS AND METHODS using Logic Pro as a case study, offer some insight into the relevance of this evolving arena to the professional, and the implications to the 16:00 enthusiast. It will conclude with some conjec - ture for the future . EB2-1 Creative Abuse in Time Stretching— Justin Engineering Brief 11 Paterson , University of West London, London, UK

34 Audio Engineering Society 130th Convention Program, 2011 Spring 16:00 derived to yield comparison charts for the 3 topologies. EB2-4 Automated Pure-Tone Audiometry Software Engineering Brief 14 Tool with Extended Frequency Range— Thomas Bisitz ,1 Andreas Silzle 2 16:00 1HörTech GmbH, Oldenburg, Germany 2Fraunhofer Institute for Integrated Circuits IIS, EB2-7 Signal Processing Applications for Erlangen, Germany Automotive Audio— Robert Cadena , Visteon In research institutes and other areas it is neces - Corporation, Van Buren Township, MI, USA sary to check the hearing of test subjects for lis - The automotive environment presents unique tening tests (e.g., for assessing the quality of challenges to audio playback. Modern technolo - audio processing algorithms, etc.) by measuring gy such as smart phones and high-efficiency pure tone audiograms. Often, neither a neces - power trains have created a new generation of sary skilled operator nor a special audiometer for audio sources and the need for audio systems high frequencies (>8kHz, important for the evalu - capable of dealing with them. This Engineering ation of a lot audio processing algorithms) are Brief will introduce emerging automotive audio available. As an alternative, a software for auto - algorithms (listed below) and their design mated audiogram measurements is presented considerations. here. This new kind of software runs on a stan - Engineering Brief 15 dard PC with a high-quality sound card and audi - ological headphones and is operated by the test subject him/herself (self-screening). The imple - Sunday, May 15 16:00 Room Saint Julien mented adaptive procedure allows fast standard Technical Committee Meeting on Network Audio audiogram measurements also for high frequen - Systems cies up to 16 kHz. The challenges with respect to dynamic range and calibration are discussed. Engineering Brief 12 Workshop 14 Sunday, May 15 16:30 – 18:00 Room 2 16:00 CHAMPAGNE LIFESTYLE ON A BEER BUDGET EB2-5 A Simple Reliable Power Amplifier with Minimal Component Count— John Chairs: Simon Bishop , UK Vanderkooy , Kevin B. Krauel , University Richard Merrick , UK of Waterloo, Waterloo, Ontario, Canada Simon Bishop and Richard Merrick contrast location We study an audio power amplifier that has audio acquisition from both ends of the budget spectrum. three essential active components: two power Discussing and comparing techniques and tricks collec - MOSFETS and one operational amplifier. Such tively acquired from 60 man-years of experience, from an amplifier will be reliable because MOSFETS being awash with money and equipment to begging, bor - have good safe-operating area properties, and rowing, and hunting on eBay. Light-hearted, but informa - there are no small semiconductors that require tive, both will prove that it’s not the size of your nail, but high voltage ratings. The topology is that of an the skill of the guy with the hammer! op-amp directly driving a grounded-source com - Regardless of your budget, we’ll prove you still have to plementary Class-B MOSFET output stage. The have STANDARDS!! center-tapped power supply for such an amplifier is floating, so each channel must have a sepa - Student/Career Development Event rate supply, and there must be a small ±15-volt RECORDING COMPETITION—PART 2 supply for the op-amp as well. We discuss the design and study the amplifier with simulations Sunday, May 15, 16:30 – 18:30 and an experimental prototype. It achieves good Room 3 performance. The Student Recording Competition is a highlight at each Engineering Brief 13 convention. A distinguished panel of judges participates in critiquing finalists of each category in an interactive 16:00 presentation and discussion. This event presents stereo and surround recordings in these categories: EB2-6 Parametric Study of Magnet System Topologies for Microspeakers— Holger Hiebel , • Traditional Multi-Track Studio Recording 16:30 to NXP Semiconductors Austria GmbH, “Sound 17:30 Solutions,” Vienna, Austria • Modern Multi-Track Studio Recording 17:30 to 18:30 A parametric simulation study was done in order to compare 3 different electrodynamic magnet The top three finalists in each category, as identified by system topologies: center-magnet, ring-magnet, our judges, present a short summary of their production and double-magnet (combined inner and outer intentions and the key recording and mix techniques magnet ring). The study is based on a row of used to realize their goals. They then play their projects simulations of the BL-factor (FEM and coil wind - for all who attend. Meritorious awards are determined ing calculation), moving mass and effective radi - here and will be presented at the closing Student Dele - ating area of a microspeaker design where the gate Assembly Meeting (SDA-2) on Monday afternoon. inner coil diameter was changed. The depen - The competition is a great chance to hear the work of dency of the sound pressure level, the electrical your fellow students at other educational institutions. quality factor, and the resonance frequency in a Everyone learns from the judges’ comments even if your closed box on the inner coil diameter were project isn’t one of the finalists, and it’s a great chance to ¯ meet other students and faculty members. Audio Engineering Society 130th Convention Program, 2011 Spring 35 Technical Progra m Workshop 15 Sunday, May 15 fit the findings of our research. The used thermal model 17:00 – 18:00 Room 4 is of second order and includes nonlinearities to describe the change of the voice coil resistance and the velocity GREEN ISSUES AND THE FUTURE OF TOURING depending cooling of the voice coil. SOUND Under the test are two micro loudspeakers one with a round and one with a rectangular voice coil because the Chair: Phil Anthony , Martin Audio request for higher output and lower resonance frequen - cies have moved the development from small round coils to larger (relative to the transducer) rectangular coils. Panelists: Catherine Langabeer , Julie’s Bicycle The effect of this transition will be documented and the Claudio Lastrucci , Powersoft differences as well as the change in the thermal factors Andy Mead , Firefly Solar will be explained. We want to prove with this session that A 2007 study by the University of Oxford’s Environmental the trend has an overall positive effect on the overall Change Institute, commissioned by music industry thermal behavior of the voice coil and the micro trans - “greening” body Julie’s Bicycle, found that the annual ducer’s capability to endure higher input power. overall Greenhouse gas emissions for the UK music Finally this tutorial should end with a discussion of the industry alone were approximately half a million tons of factual situation in the micro transducer market that the carbon. Of this total, three quarters of the impact customers are asking for more power from the driving were derived from activities associated with live music amplifier and first question is; will the micro transducer be performances. able to handle that input power? Today the amplifiers With such a large share of the UK music industry’s deliver 1-1.5 W in 8 Ohms with a build in voltage step up emissions, finding environmentally friendly solutions to but customers are asking for 2-3 W and micro transducers the live sector’s carbon and environmental impacts is an will have to be designed to handle input abuse of this ongoing concern—and an opportunity to get ahead of scale. Once this power handling is in place the next ques - artist and audience demand, more volatile energy costs tion will be; what is gained on the output side of the micro with all the knock-on impacts, and emerging regulation. transducer? Phil Anthony from Martin Audio, manufacturers of pro - fessional audio sound products, will chair this session Sunday, May 15 17:00 Room Saint Julien and will also be explaining how they reduced the carbon Technical Committee Meeting on Semantic Audio footprint of operating their new touring sound system. Analysis Alongside Phil, panelists from Julie’s Bicycle, a non-profit company working with the creative industries to coordi - nate best practice in sustainability and greenhouse gas Sunday, May 15 17:00 Room Mouton Cadet emission reduction; Powersoft, the developers of efficient Standards Committee Meeting SC-04-01 on “Green Audio Power” amplifiers; and Firefly Solar, whose Acoustics and Sound Source Modeling provision of AV equipment and solar generators aims to move the industry towards sustainable power solutions, will present information on how members of the live com - munity are implementing green initiatives in their events, Special Event tours, and technology, many at low or no additional cost. BANQUET Sunday, May 15, 18:45 – 22:30 The Musical Museum 399 High Street Tutorial 9 Sunday, May 15 Brentford, Middlesex 17:00 – 18:00 Room 5 This year the Banquet will take place at The Musical POWER LIMITATIONS IN MICRO TRANSDUCERS Museum containing one of the world’s foremost collec - tions of automatic instruments. From the tiny clockwork Presenter: Bo Rohde Pedersen , Rohde Consulting Musical Box to the self playing “Mighty Wurlitzer,” the Andreas Eberhardt Sørensen , Pulse HVT collection embraces an impressive and comprehensive array of sophisticated reproducing pianos, orchestrions, Voice coil temperature is typically a HOT topic when orchestrelles, residence organs, and violin players. On discussing power handling of transducers and many the ground floor there are 4 galleries to display working manufactures use IEC tests to specify and rank the input instruments; upstairs is a concert hall seating 230, com - power a specific transducer can endure. For micro plete with stage and, of course, an orchestra pit from speakers this situation is also the case but as we are which the Wurlitzer console will rise to entertain you, just dealing with small coils and thin wires, the need to know as it did in the cinema in the 1930s. and map the thermal factors are even more important. While having a pre-dinner drink, you will be able to The factors are the input power, losses, and cooling wander round the exhibits and even listen to some of effects that happen around the voice coil. In this session them. There will also be time at the end of the evening to these factors will be mapped and investigated as well as look round if you missed it before dinner. demonstrated. The first part of the session holds a mini The ticket price includes all food and drinks and the tutorial in using an infrared thermal camera to measure bus to the museum and back. There are limited spaces the voice coil temperature as well as a discussion on at this event, so book now to guarantee a place! The bus what must be in focus when setting up such a system to will leave the Novotel at 18:45 and return approximately measure micro transducer thermal aspects. half an hour later for the second group. The bus will be As the temperature increases our tests shows temper - available at 22:30 for those needing an early night and atures of up to 200°C in the voice coil and as a result the will return to collect the final group approximately half-an- Rdc (Dc resistance) also increases and the resulting hour later. power dissipated decreases. The theoretical thermal £ 60 for AES members and nonmembers model of the transducer will be examined and revised to Tickets will be available at the Special Events desk.

36 Audio Engineering Society 130th Convention Program, 2011 Spring Session P18 Monday, May 16 09:00 – 13:00 The method requires a handclap recording that Room 1 precedes speech activity. A running kurtosis technique is applied in order to extract an esti - SOURCE ENHANCEMENT mation of the late reflections of the room impulse response from the clap while a moving average filter is employed for the noise estimation. More - Chair: John Mourjopoulos , University of Patras, over, the excitation signal derived from the Lin - Patras, Greece ear Prediction (LP) analysis of the noisy speech along with the estimated power spectrum of the late reflections are used to suppress late rever - 09:00 beration through spectral subtraction while a Wiener filter compensates for the ambient noise. P18-1 Dereverberation in the Spatial Audio Coding A gain magnitude regularization step is also 1 Domain— Markus Kallinger, Giovanni Del implemented to reduce overestimation errors. 1 1 2 Galdo, Fabian Kuech , Oliver Thiergart Objective and subjective results show that the 1 Fraunhofer Institute for Integrated Circuits IIS, proposed method achieves significant speech Erlangen, Germany enhancement in all tested cases. 2 International Audio Laboratories, Erlangen, Convention Paper 8431 Germany Spatial audio coding techniques are fundamental 10:30 for recording, coding, and rendering spatial P18-4 System Identification for Acoustic Echo sound. Especially in teleconferencing, spatial Cancellation Using Stepped Sine Method sound reproduction helps in making a conversa - Related to FFT Size— TaeJin Park, Seung Kim, tion feel more natural reducing the listening Koeng-mo Sung , Seoul National University, effort. However, if the acoustic sources are far Seoul, Korea from the microphone arrangement, the rendered sound may easily be corrupted by reverberation. A stepped sine method was applied for system This paper proposes a dereverberation tech - identification to cancel acoustic echoes in a nique, which is integrated efficiently into the speaker phone system that has been widely parameter domain of Directional Audio Coding used in recent mobile devices. We applied the (DirAC). Utilizing DirAC’s signal model we derive stepped sine method by regarding Discrete a parametric method to reduce the diffuse por - Fourier Transform (DFT) as a uniform-DFT filter tion of the recorded signal. Instrumental quality bank. By using this stepped sine method, we measures and informal listening tests confirm were able to obtain more accurate and detailed the efficiency of the proposed method to render characteristics of non-linearity, dependent on the a spatial sound scene less reverberant without amplitude and frequency of speech. We stored introducing noticeable artifacts. the non-linearity information into linear transform Convention Paper 8429 matrices and estimated the responses of the mobile device speaker. The proposed method 09:30 exhibits higher echo return loss enhancement (ERLE) and increased correlation when com - P18-2 Blind Single-Channel Dereverberation for pared to the conventional method. Music Post-Processing— Alexandros Tsilfidis, Convention Paper 8432 John Mourjopoulos , University of Patras, Patras, Greece 11:00 Although dereverberation can be useful in many P18-5 Using Spaced Microphones with Directional audio applications, such techniques often intro - Audio Coding— Mikko-Ville Laitinen ,1 Fabian duce artifacts that are unacceptable in audio Kuech ,2 Ville Pulkki 1 engineering scenarios. Recently, the authors 1Aalto University, Espoo, Finland have proposed a novel dereverberation 2Fraunhofer Institute for Integrated Circuits IIS, approach, suitable for both speech and music Erlangen, Germany signals, based on perceptual reverberation mod - eling. Here, the method is fine-tuned for sound Directional audio coding (DirAC) is a perceptual - engineering applications and tested for both nat - ly motivated method to reproduce spatial sound, ural and artificial reverberation. The results show which typically uses input from first-order coinci - that the proposed technique efficiently suppress - dent microphone arrays. This paper presents a es reverberation without introducing significant method to additionally use spaced microphone processing artifacts and the method is appropri - setups with DirAC. It is shown that since diffuse ate for the post-processing of music recordings. sound is incoherent between spatially separated Convention Paper 8430 microphones at certain frequencies, no decorre - lation in DirAC processing is needed, which 10:00 improves the perceived quality. Furthermore, the directions of sound sources are perceived to be P18-3 Joint Noise and Reverberation Suppression more accurate and stable. for Speech Applications— Elias K. Kokkinis, Convention Paper 8433 Alexandros Tsilfidis, Eleftheria Georganti, John Mourjopoulos , University of Patras, Patras, 11:30 Greece P18-6 Parameter Estimation in Directional Audio An algorithm for joint suppression of noise and Coding Using Linear Microphone Arrays— reverberation from speech signals is presented. Oliver Thiergart ,1 Michael Kratschmer, 2 Markus ¯ Audio Engineering Society 130th Convention Program, 2011 Spring 37 Technical Progra m Kallinger, 2 Giovanni Del Galdo 2 attempt to identify time-frequency atoms that 1International Audio Laboratories, Erlangen, belong to only one segment. The system is Germany described, with reference to the relevant underly - 2Fraunhofer Institute for Integrated Circuits IIS, ing theory, and the quality of its output for the best Erlangen, Germany bases from complex wavelet packets is compared with methods based on more established analysis Directional Audio Coding (DirAC) provides an and processing methods. efficient description of spatial sound in terms of Convention Paper 8436 few audio downmix signals and parametric side information, namely the Direction Of Arrival (DOA) and the diffuseness of the sound. Tradi - tionally, the parameters are derived based on Session P19 Monday, May 16 09:00 – 10:30 the active sound intensity vector that is often Foyer determined via 2-D or 3-D microphone grids. Adapting this estimation strategy to linear arrays, POSTERS: ROOM ACOUSTICS which are preferred in various applications due to form factor constraints, yields comparatively poor results. This paper proposes to replace the 09:00 intensity based DOA estimation in DirAC by a P19-1 System Identification of Equalized Room specific estimation of signal parameters via rota - Impulse Responses by an Acoustic Echo tional invariant techniques, namely ESPRIT. Canceller Using Proportionate LMS Moreover, a diffuseness estimator exploiting the Algorithms— Stefan Goetze ,1 Feifei Xiong ,1 Jan correlation between the array sensors is pre - Ole Jungmann ,2 Markus Kallinger ,3 Karl-Dirk sented. Experimental results show that the Kammeyer ,4 Alfred Mertins 2 DirAC concept can be applied in practice also in 1Fraunhofer Institute for Digital Media conjunction with linear arrays. Technology IDMT, Ilmenau, Germany Convention Paper 8434 2University of Lübeck, Lübeck, Germany 3University of Oldenburg, Oldenburg, Germany 12:00 (now with Fraunhofer Institute for Integrated P18-7 Extraction of Voice from the Center of the Circuits IIS, Erlangen, Germany) 4University of Bremen, Bremen, Germany Stereo Image— Aki Härmä, Munhum Park , Philips Research Laboratories Eindhoven, Hands-free telecommunication systems usually Eindhoven, The Netherlands employ subsystems for acoustic echo cancella - tion (AEC), listening-room compensation (LRC), Detection and extraction of the center vocal and noise reduction in combination. This contri - source is important for many audio format con - bution discusses a combined system of a two- version and manipulation applications. First, we stage AEC filter and an LRC filter to remove study some generic properties of stereo signals reverberation introduced by the listening room. containing sources panned exactly to the center An inner AEC is used to achieve initial echo re - of the stereo image and propose an algorithm for duction and to perform system identification the separation of a stereo audio signal into a needed for the LRC filter. An additional outer center and side channels. In the 128th AES AEC is used to further reduce the acoustic Convention a paper was presented (Convention echoes. The performance of proportionate filter Paper 8071) on listening tests comparing the update schemes such as the so-called propor - perception of the widths of the stereo images of tionate normalized least mean squares algorithm synthetic signal combinations. In this paper the (PNLMS) or the improved PNLMS (IPNLMS) for same experiment is repeated with real stereo system identification of equalized impulse audio content using the proposed center separa - response (IR) are shown and the mutual influ - tion algorithm. The main observation is that ences of the subsystems are analyzed. If the there are clear differences in the results. The LRC filter succeeds in shaping a sparse overall reasons for the differences are discussed in light IR for the concatenated system of LRC filter and of the literature and analysis of the test signals room impulse response (RIR) the PNLMS per - and their binaural characteristics in the listening forms best since it is optimized for the identifica - test setup. tion of sparse IRs. However, the equalization Convention Paper 8435 may be imperfect due to channel estimation er - rors in periods of convergence and due to the 12:30 so-called tail-effect of AEC, i.e., the fact that only P18-8 Directional Segmentation of Stereo Audio via the first part of a RIR is identified in practical Best Basis Search of Complex Wavelet systems. The IPNLMS is more appropriate in Packets— Jeremy Wells , University of York, this case to identify the equalized IR. York, North Yorkshire, UK Convention Paper 8437

A system for dividing time-coincident stereo 09:00 audio signals into directional segments is present - ed. The purpose is to give greater flexibility in the P19-2 Virtual Room Acoustics : A Comparison of presentation of spatial information when two-chan - Techniques for Computing 3-D-FDTD nel audio is reproduced. For example, different in - Schemes Using CUDA— Craig Webb, Stefan ter-channel time shifts could be introduced for seg - Bilbao , University of Edinburgh, Edinburgh, ments depending on their direction. A novel aspect Scotland, UK of this work is the use of complex wavelet packet High fidelity virtual room acoustics can be analysis, along with “best basis” selection, in an

38 Audio Engineering Society 130th Convention Program, 2011 Spring approached through direct numerical simulation Loudspeaker Setups— Javier Gomez ,1 of wave propagation in a defined space. Three- Rafael C. D. Paiva ,1,2 Kai Saksela ,1 Thomas dimensional Finite Difference Time Domain Svedström ,1 Ville Pulkki 1 schemes can be employed and adapt well to a 1Aalto University, Espoo, Finland parallel programming model. This paper exam - 2Nokia Technology Institute INdT, Brazil ines the various approaches for calculating these schemes using the Nvidia CUDA architec - A listening test was conducted to assess the ture. We test the different possibilities for struc - effect that different direct-to-reverberant ratios, turing computation, based on the available loudspeaker setups, and reverberation times memory objects and thread-blocking model. A have on the directional perception of a synthetic standard test simulation is computed at double room response. The results show that the per - precision under different arrangements. We find ceived spatial distribution of reverberation is that a 2-D extended tile blocking system, com - dependent on the speaker setup, on the rever - bined with shared memory usage, produces the beration time, and on the listener. Additionally, fastest computation for our scheme. However, they show that a small amount of direct sound, shared memory usage is only marginally faster even when barely masked by reverberation, than direct global memory access, using the lat - shifts the overall perception to the front. est FERMI GPUs. Convention Paper 8441 Convention Paper 8438 09:00 09:00 P19-6 Modal Analysis and Sound Field Simulation P19-3 Acoustic Parameters of Chosen Orthodox of Vibration Panels in a Free Sound Field Churches Overview and Preliminary and in a Rectangular Enclosure— Kazuhiko Psychoacoustic Tests Using Choral Music— Kawahara, 1 Akihiro Sonogi ,1 Shin-ya Sato 1,2 1 Pawel Maleck, Jerzy Wiciak , AGH University Kyushu University, Fukuoka, Japan 2 of Science and Technology, Kraków, Poland Nittobo Acoustic Engineering Co., Ltd., Tokyo, Japan A lot of acoustic research was done for Roman Catholic churches and because of some differ - Several types of implementation of panel loud - ences in traditions and culture comparing to speakers, for example, distributed mode loud - Orthodoxy, it is not the best idea to use its speaker (DML), are proposed. The diaphragm of acoustic estimators for Orthodox churches. The a DML is thought to behave like a bending plate. paper shows results of measurements in some The diaphragm of an electrostatic loudspeaker is Orthodox churches in Poland and the proposal thought to move as a rigid plate. What is the dif - of psychoacoustic tests using the convolution ference of an acoustical sound field generation technique, which would allow formulating the feature with each implementation? In this paper new acoustic outlines for Orthodox churches. we used models of same dimensional rectangu - The research has been done especially consid - lar panel. One is a rigid diaphragm. Another is a ering choral music, which is an inseparable part bending panel that has an actuator in the center of Eastern Christian culture; so as a test, there of it. We made simulations of directivity and were Orthodox choir sound samples recorded in sound pressure distribution in a rectangular an anechoic chamber convoluted with impulse room. We made simulations of sound pressure response of measured churches. distribution in a rectangular room. We could Convention Paper 8439 show less coherent SPL distribution in the case of a bending panel. Convention Paper 8442 09:00 P19-4 A Comparative Study of Various “Optimum” 09:00 Room Dimension Ratios— John Sarris , Are - taieio University Hospital, Athens, Greece P19-7 Simple Room Acoustic Analysis Using a 2.5 Dimensional Approach— Patrick Macey , Various “optimum” room dimension ratios that PACSYS Limited, Nottingham, UK have been proposed in the literature are studied and compared. Since each proposal is based on Cavity modes of a finite bounded region with a different criterion, independent objective mea - rigid boundaries can be used to compute the sures of the acoustic quality of the various room steady state harmonic response for excitation of ratios are applied in this paper. The most an acoustic cavity by a point source. In Cuboid straightforward metric is the flatness of a room’s domains, with analytical modes, this is straight - corner to corner frequency response, but since forward. For more general regions, determining this is not representative of the variations of the a set of orthonormal modes is more difficult. The sound pressure within the closed space, differ - finite element method could be used for arbitrary ent metrics are employed to quantify these varia - 3-D regions. The current work investigates a tions. Simulation results are presented to evalu - hybrid numerical/analytical method, applicable to ate the effectiveness of the various ratios for the rooms of constant height, but arbitrary cross case of a small and a larger room. section. A 2-D FE analysis is used to compute Convention Paper 8440 the cross-section modes/frequencies. Three- dimensional modes are constructed as an outer product of the cross section modes and 1-D 09:00 modes for the height. Comparison is made with P19-5 Perception of Spatial Distribution of Room BEM computations. Response Reproduced Using Different Convention Paper 8443 ¯

Audio Engineering Society 130th Convention Program, 2011 Spring 39 Technical Progra m 09:00 recordings, speeds, equalization curves; (6) A bag of miscellaneous horrors such as color codes, connectors, P19-8 Early Reflections Design for Natural Stereo and so on. Sound Listening— Chul-Jae (Jay) Yoo , RADSONE, Inc., Gyeonggi-do, Korea Tutorial 11 Monday, May 16 For natural stereo sound in rooms, some consid - 09:00 – 10:15 Room 5 erations must be followed. For example, to maxi - mize time density, Feedback Delay Network ROADIE—HOW TO GAIN AND KEEP A CAREER (FDN) is used in which the feedback matrix has IN THE LIVE MUSIC INDUSTRY every element consisting of absolute value 1. To model faster high frequency decay according to Presenter: Andy Reynolds , Owner, the increasing number of reflections, absorption LiveMusicBusiness.com filters after delay lines in the FDN are generally used. Mixing matrix for uncorrelated L/R output Live music is a huge industry—65 million people around is also used. In this paper an elaborate design the world attended a concert in 2010, with gross ticket scheme of the first early reflection portion in ear - sales of over $3 billion. Every single one of those con - ly reflections was proposed to obtain higher certs needs a sound reinforcement system and audio IACC values than those of conventional sys - engineers to rig the system and help enhance the sound tems, resulting in a more precise sound source coming from stage. image with ambience. But how do you gain, and more importantly keep, an Convention Paper 8444 audio job in the live music business? How do you differ - entiate between the different audio roles on a show or event? What specialized skills are relevant to concert Workshop 16 Monday, May 16 audio engineering? 09:00 – 11:00 Room 3 This tutorial will examine the role of the audio engineer in modern concert production, look at routes into the in - dustry, offer advice on finding and dealing with clients as UP AND DOWNMIXING—APPROACHES, SOLUTIONS well as the technical skills and knowledge that every live audio engineer should be aware of. Chair: Florian Camerer , ORF – Austrian TV Special Event Panelists: Tom Allom , Penteo AES/APRS/DV247 EVENT: TALKBACK-PRO 3 Antoine Hurtado , Isostem THE MYTHS AND MYSTERIES OF DSP Clemens Par , Swissaudec Monday, May 16, 09:00 – 10:45 Pieter Schillebeeckx , Soundfield Room 2 Christian Struck , Lawo Moderator: Wes Maebe Upmix solutions are coming to the market with increasing speed. Broadcasters need those solutions especially for Where would we be without DSP? Given the colossal HD services to fulfill the expectations of the audience for and ubiquitous influence DSP processing has on all our consistent surround sound presentation. Various functionality, the answer probably would be “way back in approaches can be found that try to provide a convincing the analog world” but is there good DSP and bad DSP? illusion of a discrete surround sound signal. In this work - What should our priorities be when it comes to spending shop, 6 company representatives will present their prod - on A/D – D/A, soft auxiliaries, tracking and mixing pro - ucts in a short introduction. Then their upmix results of 2 grams, and master/finalizing tools. stereo tracks previously given to them will be played and Issues with interfaces, latency, compression, and plain discussed. Also, downmix aspects will be touched as this is taste will always plague the subjective judgments of also a vital ingredient of a suitable surround sound signal. engineers and producers. Is there a “normal” digital base upon which creativity can grow unhindered by the pitfalls Tutorial 10 Monday, May 16 of mismatched technologies, uneven standards, and de - 09:00 – 10:30 Room 4 liberate hurdles to interoperability. Clearing the cloudy mists that envelop proprietary toys will be a panel of technologists, enthusiasts, and ANALOG STANDARDS—WHY ARE THERE SO practitioners. MANY? Monday, May 16 09:00 Room Saint Julien Presenter: Sean Davies , S.W. Davies Ltd., Aylesbury, UK Standards Committee Meeting AESSC Plenary “We like standards, that’s why we have so many.” (variously attributed) Tutorial 12 Monday, May 16 10:30 – 12:00 Room 5 Although most current audio work is in the digital domain there remains a vast legacy of recordings made in ana - YOU, A ROOM, AND A PAIR OF HEADPHONES: log, access to which will be required for some consider - A LESSON IN BINAURAL AUDIO able time to come. Knowledge is therefore required of the standards applicable to each recording, also the rele - vant technical literature. Presenter: Ben Supper , Focusrite This tutorial will seek to explain how and why the Stereo recordings that are intended for loudspeaker seemingly perverse plurality of standards arose. Fields reproduction do not work properly over headphones. covered will be: (1) Impedances; why 30 ohms (“R”), Methods for converting stereo for headphone monitoring 200R, 300R, 600R; (2) Levels; the TU, the Neper, the have become increasingly elaborate as the availability of Decibel; (3) Metering, the VU versus the PPM; (4) Disc digital signal processing has increased. The challenge is recordings; speeds, equalization curves; (5) Tape 40 Audio Engineering Society 130th Convention Program, 2011 Spring to arrive at a stereophonic effect that is as the composer, tant for designing small and light transducers producing engineer, and producer intended: stable, with correct the desired output at high efficiency and reasonable cost. perspective, out-of-head localization, and no excessive polarization of the stereo image. Workshop 18 Monday, May 16 To do these things successfully is not as trivial as may 11:30 – 13:00 Room 3 first be assumed. No two listeners are identical; choice of headphones are a matter of taste. Moreover, the illusion THE EUROVISION SONGCONTEST 2010 IN OSLO needs reverberation, but this must not get in the way of —A CASE STUDY the music. This tutorial picks a path through the peculiarities of Gaute Nistov the human hearing system, discussing aspects of our Co-chairs: , NRK Oslo Ina von Lucas perception of loudspeakers and rooms, and how close , NRK Oslo we can come to understanding and modeling them com - The Eurovision Songcontest is the biggest live transmis - pletely. Insights will be given into how the human audito - sion in Europe. The demands on all aspects of produc - ry system can be made to believe a virtually-rendered tion are daunting and the bar has been raised in past sound field, how to do this cheaply, and why we don’t years to challenging heights. Sound Supervisor Gaute need to cram the entire gamut of reality onto a tiny piece Nistov tells how he stepped up to the plate, and Ina van of silicon. Lucas will elaborate on the complex communication and routing system installed. Monday, May 16 10:30 Room Mouton Cadet Technical Committee Meeting on Fiber Optics Monday, May 16 12:30 Room Mouton Cadet for Audio Technical Council Meeting

Workshop 17 Monday, May 16 Session P20 Monday, May 16 14:00 – 16:00 11:00 – 13:00 Room 2 Room 1

(WHY) DOES LIVE SOUND HAVE A BAD NAME? SUBJECTIVE EVALUATION

Chair: Mark Bailey , UK Chair: Bruno Fazenda , University of Salford, Salford, UK Panelists: Nichola Bailey , Nichola Bailey School of Voice Bob Lee , QSC Audio Products David Millward , Freelance FOH Engineer 14:00 Ian Nelson , FOH Engineer, Adlib Audio Bruce Olson , Acoustic Consultant, ADA P20-1 Selection of Audio Stimuli for Listening Tests 1 1 2 Acoustics — Jonas Ekeroot, Jan Berg, Arne Nykänen 1 Tuomo Tolonen , Shure Distribution UK Luleå University of Technology, Piteå, Sweden 2Luleå University of Technology, Luleå, Sweden In this session we will explore the aspects that contribute to good sound and those that conspire against it. Our Two listening test methods in common use for panel of practitioners, users and manufacturers will dis - the subjective assessment of audio quality are cuss and debate matters determined to be the scourge the ITU-R recommendations BS.1116-1 for small (and saviour) of live audio. impairments and BS.1534-1 (MUSHRA) for inter - In addition to discussion, we will attempt to practically mediate quality. They stipulate the usage of only demonstrate some of the points. We look forward to see - critical audio stimuli (BS.1116-1) to reveal differ - ing you there! ences among systems under test, or critical au - dio stimuli that represents typical audio material in a specific application context (MUSHRA). A Tutorial 13 Monday, May 16 poor selection of stimuli can cause experimental 11:00 – 13:30 Room 4 insensitivity and introduce bias, leading to incon - clusive results. At the same time this selection HOT AND NONLINEAR—LOUDSPEAKERS process is time-consuming and labor-intensive, AT HIGH AMPLITUDES and is difficult to conduct in a systematic way. This paper reviews and discusses the selection Presenter: Wolfgang Klippel , Klippel GmbH, Dresden, of audio stimuli in listening test-related studies. Germany Convention Paper 8445

Nonlinearities inherent in electro-dynamical transducer 14:30 and the heating of the voice coil and magnetic system limit the acoustical output, generate distortion and other P20-2 A Listening Test System for Automotive symptoms at high amplitudes. The large signal perfor - Audio: PART 5 – The Influence of Listening mance is the result of a deterministic process and Environment on the Realism of Binaural predictable by lumped parameter models comprising Reproduction— Francois Postel ,1,2 Patrick nonlinear and thermal elements. The tutorial gives an in - Hegarty ,3 Søren Bech 3 troduction into the fundamentals, shows alternative mea - 1Bang & Olufsen A/S, Struer, Denmark (now at surement techniques and discusses the relationship Arkamys, Paris, France) between the physical causes and symptoms depending 2Department of Mechanical Vibrations and on the properties of the particular stimulus (test signal, Acoustics, UTC, Compiegne, France music). Selection of meaningful measurements, the inter - 3Bang & Olufsen A/S, Struer, Denmark pretation of the results and practical loudspeaker diag - Binaural technology is used to capture elements nostic is the main objective of the tutorial, which is impor - of an in-car sound field and reproduce them over ¯ Audio Engineering Society 130th Convention Program, 2011 Spring 41 Technical Progra m headphones at another place and time. An designed to assess the subjective performance experiment to test the influence of the listening of a selection of upmixers for use in the broad - environment on the realism of such a binaural cast chain. reproduction is described. A panel of 12 trained Convention Paper 8448 listeners rated a range of stimuli for 6 elicited attributes of sound quality. Ratings are made for the actual sound field in the test vehicle, for a Student/Career Development Event binaural reproduction in the same vehicle, and STUDENT DELEGATE ASSEMBLY MEETING— for a binaural reproduction in a listening room. PART 2 The results show that the tested binaural repro - Monday, May 16, 12:15 – 13:45 duction system is able to preserve either the Room 5 rank order or the perceived magnitudes of the impressions of the sound field for the attributes At this meeting the SDA will elect a new vice chair. One Precision, Treble, Stereo impression, Bass, and vote will be cast by the designated representative from Reverberation, independent of the listening each recognized AES student section in the Europe and environment. International Regions. Judges’ comments and awards Convention Paper 8446 will be presented for the Recording Competitions. Plans for future student activities at local, regional, and interna - tional levels will be summarized. 15:00 P20-3 Differences in Preference for Noise Reduction Strength between Individual Session P21 Monday, May 16 14:00 – 15:30 Listeners— Rolph Houben ,1 Tjeerd M. H. Foyer Dijkstra ,2,3 Wouter A. Dreschler 1 1Academic Medical Center, Amsterdam, POSTERS: PROCESSING AND ANALYSIS The Netherlands 2Radboud University, Nijmegen, The Netherlands 3Technical University Eindhoven, The 14:00 Netherlands P21-1 Automatic Classification of Musical Audio There is little research on user preference for dif - Signals Employing Machine Learning ferent settings of noise reduction, especially for Approach— Pawel Zwan, Bozena Kostek, Adam individual users. We therefore measured individ - Kupryjanow , Gdansk University of Technology, ual preference for pairs of audio streams differ - Gdansk, Poland ing in noise reduction strength. Data was ana - lyzed with a logistic probability model that is This paper presents a thorough analysis of auto - based on a quadratic preference utility function. matic classification applied to musical audio sig - This allowed for an estimate of the optimal set - nals. The classification is based on a chosen set ting for each individual subject. For five out of of machine learning algorithms. A database of ten subjects the optimized setting differed signifi - 60 music composers/performers was prepared cantly from the optimum obtained for the for the purpose of the described research. For grouped data. However, the predicted prefer - each of the musicians, 15 to 20 music pieces ence for the individual optimum (60%) was only were collected. All the pieces were partitioned slightly higher than chance level (50%), which into 20 segments and then parameterized. The can be considered as too weak to advocate feature vector consisted of 171 parameters, individualization of noise reduction for these nor - including MPEG-7 low-level descriptors and mel- mally hearing subjects. However, in hearing- frequency cepstral coefficients (MFCC) comple - impaired subjects this may be different. mented with time-related dedicated parameters. Convention Paper 8447 The task of the classifier was to recognize the composer/performer and to properly categorize a selected piece of music. The paper also pre - sents and discusses the results of classification. 15:30 Convention Paper 8449 P20-4 Assessment of Stereo to Surround Upmixers for Broadcasting— David Marston , BBC R&D, 14:00 London, UK P21-2 Evaluation of Onset Detection Algorithms Broadcasters are now transmitting 5.1-channel in Popular Polyphonic Music on a Large surround sound as part of their HD TV services. Scale Database— Stephan Hübler, Rüdiger However, since much of the audio content in Hoffmann , Technische Universität Dresden, original program material is 2-channel stereo, Dresden Germany broadcasters are required to switch between the two formats. Switching between 2- and 5.1- This paper introduces a large database of popu - channel formats can cause problems in decod - lar polyphonic music containing drums (10,238 ing the content, including switching artifacts and onsets) for the evaluation of onset detection loudness changes. In general it is preferable to algorithms. The database has been manually transmit all program audio in 5.1 surround, and annotated by expert listeners. The inter-rater this can be achieved by automatically upmixing variability leads to an understanding of inter- any 2-channel stereo content to 5.1 format prior human variations. Four common detection func - to broadcasting. This paper reports on tests tions are investigated: spectral difference, high

42 Audio Engineering Society 122nd Convention Program, 2007 Spring frequency content, phase deviation, and the psy - 14:00 choacoustic one of Klapuri. We present an addi - P21-5 Selection of Approximated Activation tional detection function based on the MPEG7 Functions in Neural Network-Based Sound feature audio spectrum envelope. An adaptive Classifiers for Digital Hearing Aids— peak picker determines the onsets that are com - Lorena pared with the manual labels. Results show that Álvarez, Cosme Llerena, Enrique Alexandre, detection functions based on spectral difference Roberto Gil-Pita, Manuel Rosa-Zurera , obtain observable better results. The study pro - University of Alcalá, Alcalá de Henares, Spain vides a thorough investigation of onset detection The feasible implementation of signal processing algorithms in popular polyphonic music. techniques on hearing aids is constrained by the Convention Paper 8450 limited number of instructions per second to [Paper presented by Till Moritz Essinger] implement the algorithms on the digital signal processor the hearing aid is based on. This adversely limits the design of a neural network- 14:00 based classifier embedded in the hearing aid. Aiming at helping the processor achieve accu - P21-3 Using the Viterbi Algorithm for Error rate enough results, and in the effort of reducing Correction in an Autocorrelation-Based Pitch the number of instructions per second, this Detector— Bob Coover , Gracenote, Inc., paper focuses on exploring the most adequate Emeryville, CA, USA approximations for the activation function. The An autocorrelation-based method for detecting experimental work proves that the approximated the fundamental pitch of an audio signal is pre - neural network-based classifier achieves the sented in which the Viterbi algorithm is used in same efficiency as that reached by the “exact” place of the error correction portion of the detec - networks (without these approximations), but, tor. The Viterbi algorithm is used to locate the this is the crucial point, with the added advan - most likely pitch path through the audio file. This tage of extremely reducing the computational method is compared to a typical heuristic and cost on the digital signal processor. median filtering-based error correction approach Convention Paper 8453 that has been historically used in this type of algorithm. The Viterbi algorithm results are sig - 14:00 nificantly better than the typical error correction methods for choosing the best and most plausi - P21-6 Development of Multiband Dynamic Range ble path through the pitch estimates. Compressor Regarding Noise Characteristics Convention Paper 8451 — Hoon Heo, Mingu Lee, Seokjin Lee, Koeng- Mo Sung , Seoul National University, Seoul, Korea 14:00 It is hard to hear sounds from digital TVs or P21-4 Spectral Equalization for GHA-Applied mobile phones in noisy environments because of Restoration to Damaged Historical 78 rpm the masking effect. It could be solved by a sim - Records— Teruo Muraoka, Takahiro Miura, ple amplification; however, special process for Tohru Ifukube , The University of Tokyo, Tokyo, masking bands will be a further solution in some Japan restricted situation. We proposed an algorithm named “perceptual irrelevant component elimi - The authors have been engaged in the research nation” using a modified multiband dynamic of In-harmonic Frequency Analysis “GHA,” which range compressor, which does not increase the enables the separation of desired signal-compo - signal level and enhances its perceptual signal- nents and noise. Its primary purpose has been to-noise ratio by about 1 dB for speech signals noise-reduction. Recently, the authors succeed - and about 3 dB for music signals. ed in conducting GHA in practical time length Convention Paper 8454 and carried out many sound restorations of his - torical 78 rpm records. Thanks to GHA’s suffi - cient separation of target signal-component from 14:00 noisy objects, the restored signal is noise-less, however its tone quality is unnatural when it is P21-7 Designing Sets of N Doubly Complementary reproduced using current audio equipment. This IIR Filters— Alexis Favrot, Christof Faller , Illu - is due to fact that the recorded sounds were sonic LLC, Lausanne, Switzerland tuned to match to audio equipment in that age, A filter design procedure is described for obtain - therefore spectral equalization is necessary. In ing sets of N doubly complementary IIR filters for practice, extreme frequency emphases are any N and any bandpass frequencies. The N required, but it had been impossible because of doubly complementary IIR filters are built follow - the existences of scratch noise. GHA-applied ing a tree-like structure based on pairs of doubly restoration removed theses difficulties, and complementary IIR filters and additional all-pass equalization curve was obtained by comparing filters. The sum of all band signals is an all-pass long-term spectrum of restored music with that filtered version of the original audio signal. The of the same recorded music by current musi - complementary IIR filters can be used instead of cians. Generally equalizations are very compli - an analysis filterbank (full rate). The correspond - cated and were done utilizing a parametric ing synthesis filterbank is simply the sum of all equalizer. band signals. The proposed filters enable high Convention Paper 8452 quality delay critical audio processing. Convention Paper 8455 ¯

Audio Engineering Society 122nd Convention Program, 2007 Spring 43 Technical Progra m Workshop 19 Monday, May 16 Workshop 20 Monday, May 16 14:00 – 15:30 Room 4 15:45 – 17:45 Room 4

AUTOMIXING AND ARTIFICIAL INTELLIGENCE LOUDNESS IN BROADCASTING—EBU RECOMMENDATION R 128 STEPS ON THE GAS Chair: Michael “Bink” Knowles , Independent Live Sound Mixer Chair: Florian Camerer , ORF – Austrian TV, and Chairman of EBU PLOUD Panelists: Alice Clifford , Queen Mary University of London, London, UK Panelists: Yannick Dumartineix , TSR Geneva Ron Lorman , Busy Puppy Productions Matthieu Parmentier , France Television Josh Reiss , Queen Mary University of Alessandro Travaglini , Fox Italia London, London, UK The EBU group PLOUD has now published all their This workshop will cover two aspects of automixing: the intended documents, centering around the loudness rec - current use of automixers for live sound mixing, and a ommendation R 128. Implementation of this concept is projection of the future of artificial intelligence and auto - underway, and the loudness normalization paradigm is matic functions applied to live sound. gaining ground. The end of the “loudness war” is nigh! At The first part of the workshop will focus on the prob - least in broadcasting that seems to be the prospect. . . . lems facing the operator who is mixing panel discussions In this workshop, members of PLOUD who already and improvisational theater, and the benefits of automix - implemented the concepts of R 128 will talk about their ers under the control of human hands. The second part experiences and illustrate it with examples. will consider the aspects of live sound mixing that may be totally automated in the future, including the setting of Tutorial 15 Monday, May 16 input gain, high-pass filtering, channel equalization, and 16:00 – 17:30 Room 2 the more distant goal of musical balance. DELAY FX—LET'S NOT WASTE ANY MORE TIME Tutorial 14 Monday, May 16 14:00 – 15:30 Room 3 Presenter: Alex Case , University of Massachusetts at Lowell, Lowell, MA, USA SURROUND SOUND FORMATS The humble delay is an essential effect for creating the Presenter: Hugh Robjohns , Technical Editor and loudspeaker illusion of music that is better than real, big - Trainer, Project Manager ger than life. A broad range of effects—comb filtering, flanging, chorus, echo, reverb, pitch shift, and more—are After many years of Surround Sound becoming main - born from delay. One device, one plug-in, and you’ve got stream in the broadcast business, this tutorial looks back a vast pallet of production possibilities. Join us for this at the basics of formats and technical details of surround thorough discussion of the technical fundamentals and sound, covering diverse matrix techniques like Dolby production strategies for one of the most powerful signal Surround, compression systems like DTS or Dolby AC-3, processes in the audio toolbox: delay. as well as broadcast-specific coding schemes like Dolby E. Metadata, lineup tones, and mixing issues with respect to the coding systems will be touched upon, too.

Special Event AES/APRS EVENT: I DIDN'T GET WHERE I AM TODAY . . . Monday, May 16, 14:00 – 15:45 Room 2 Moderator: Peter Filleul , APRS Panelists: David Fisher Elliott Randall Dennis Weinreich Careers are magically unpredictable things. So many successful icons in the audio world will blame serendipi - ty, a chance meeting, or the whims of luck for how they ended up. Others had a clear direction, a specific focus, and have stuck to their original ambition and some swear by training and qualifications. Prominent audio characters describe what influenced their “career moves” and whether flexibility, a broad out - look and the “maverick” spirit are essential elements to achieving longevity in the audio business and muse on the expectations generated by audio education and training. A 90 minute seminar moderated by Peter Filleul of the APRS exploring the differences between the theory and the practice of the career ambitions of a set of accom - plished audio addicts.

44 Audio Engineering Society 130th Convention Program, 2011 Spring