DOCSLIB.ORG

Handout Lecture / Tutorial on Music Information Retrieval

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Handout Lecture / Tutorial on Music Information Retrieval Lead-in Music Information Retrieval Who am I? Vienna University of Technology http://www.tuwien.ac.at http://www.ifs.tuwien.ac.at/mir • Faculty of Computer Science http://www.cs.tuwien.ac.at – Department of Software Technology and Interactive Systems Andreas Rauber http://www.isis.tuwien.ac.at » Software and Information Engineering Group Department of Softwaretechnology and http://www.ifs.tuwien.ac.at Interactive Systems - Andreas Rauber Vienna University of Technology http://www.ifs.tuwien.ac.at/~andi http://www.ifs.tuwien.ac.at/~andi Machine Learning, Neural Networks Text Mining, Digital Libraries Music Retrieval Digital Preservation . Lead-in Lead-in Activities Who else is MIR@ifs? Audio Feature Extraction Thomas Lidy Music Classification Robert Neumayer PlaySOM: Organisation of Music Archives PocketSOM: Browsing Music on Mobile Devices Rudolf Mayer 3D Worlds for Music Jakob Frank Audio Segmentation Chord Detection Other members Former members Blind Source Separation Veronika Zenz Markus Frühwirth Text and Music (Lyrics, Bio, ...) Peter Hlavac Elias Pampalk Ewald Peiszer Stefan Leitich Andreas Scharf David Laister Andrei Grecu & Doris Baum & others others . Chorus Music IR – Music? What is „Music“? Lead-in Music, of course! Chorus Audio: wav, au, mp3, ... Verse 1: Music-IR Symbolic: MIDI, mod, ... Verse 2: Audio Features www.samplesmith.com Scores: Scan, MusicXML Verse 3: Classification and Benchmarking www.westminster.gov.uk Verse 4: Clustering & Browsing Text Community data Video/Images Verse 5: Some other applications – Song lyrics – Playlists – Album covers Fade-out – Artis Biographies – Market basket – Music videos – Websites: – Band evolution Fanpages, Album Reviews, Genre descriptions . 1 Music IR – Music? Music IR – Music? Music - Sound Music - Sound - Loudness http:// www.phys.unsw.edu.au/jw/hearing.html Sound as acoustic wave Source of sound sound pressure sound pressure level Characterized by the properties of waves pascal dB re 20 µPa immediate soft tissue damage 50000 approx. 185 (frequency/wavelength, amplitude) threshold of pain 100 134 Frequency: pitch hearing damage during short-term effect 20 approx. 120 jet engine, 100 m distant 6–200 110–140 – Humans can hear approx. 20Hz-20kHz jack hammer, 1 m distant / discotheque 2 approx. 100 – speech: 200Hz-8kHz hearing damage during long-term effect 0.6 approx. 85 major road, 10 m distant 0.2–0.6 80–90 Amplitude: Loudness passenger car, 10 m distant 0.02–0.2 60–80 – measured as pressure in micropascal µPa TV set at home level, 1 m distant 0.02 ca. 60 normal talking, 1 m distant 0.002–0.02 40–60 – hearing threshold: approx. 20 µPa very calm room 0.0002–0.0006 20–30 – logarithmic decibel scale leaves noise, calm breathing 0.00006 10 auditory threshold at 2 kHz 0.00002 0 . Music IR – Music? Music IR – Music? Music - Sound Music - Sound Nyquist sampling theorem: Different file formats for storing sound: Exact reconstruction of a continuous-time baseband signal from its – lossless formats samples is possible if the signal is bandlimited and the sampling • WAV (may hold compressed audio, but usually lossless PCM) frequency is greater than twice the signal bandwidth. • FLAC, Shorten, Monkey's Audio, ATRAC Advanced Lossless, Apple Lossless, WMA Lossless, TTA is the Nyquist frequency, i.e. a signal with a specific frequency – lossy formats must be sampled with twice that frequency for reconstruction. • MP3 • ATRAC More on sound, sound pressure, hearing thresholds, etc. later when • AAC we talk about feature extraction from sound. • Ogg Vorbis • WMA • ... Music IR – Music? Music IR – Music? Music - Sound - PCM Music - Sound - MP3 PCM: Pulse Code Modulation Actually: MPEG-1 Audio Layer 3 Digital representation of an analog signal where the magnitude of Developed by a groups around Fraunhofer, Thomson, the signal is sampled regularly at uniform intervals, then quantized AT&T Bell Labs, several patent issues pending to a series of symbols Lossy compression, based on psycho-acostic models Used in WAV, CD-recordings, ... – differential encoding of stereo signal (lossless) Quantization error: chosing discrete value near the analog signal – focus on audible frequencies for each sample – masking effects Any frequency above or equal to – adaptive bit-depth encoding 1/2 sampling frequency is lost – quantization and huffman-encoding . 2 Music IR – Music? Music IR – Music? Music - Sound - MP3 What is „Music“? ID3-Tags Music, of course! Added later-on to allow embedding of meta data – Audio: wav, au, mp3, ... – Symbolic: MIDI, mod, ... ID3v1: 30 char per entry, few standard fields www.samplesmith.com – Scores: Scan, MusicXML ID3v2.4: UTF-8 support, tags at beginning of file www.westminster.gov.uk Used by search engines Text Community data Video/Images – Song lyrics – Playlists – Album covers – Artis Biographies – Market basket – Music videos – Websites: – Band evolution Fanpages, Album Reviews, Genre descriptions . Music IR – Music? Music IR – Music? Musical Instrument Digital Interface - MIDI Musical Instrument Digital Interface - MIDI Some MIDI examples Symbolic Music File Format (from: http://www.borg.com/~jglatt/files/midifile.htm ) Dave Smith, proposed in 1981 – Orchestral: Bach: Branderburg Concerto 4 – Orchestral: Star Treck Theme: Next Generation MIDI specification 1.0 in 1983 – Classic: Beethoven: Für Elise Interacting with keyboard produces messages – 1950's Rock&Roll: Bill Haley: Rock Around the Clock – 1950's Rock&Roll: Jerry Lee Louis: Great Balls of Fire – Note-On , Aftertouch , and Note-Off – Pop: Elton John: Don't Let the Sun Go Down – 127 note pitches – Pop: Phil Colins: Another Day in Paradise Sequence of control commands – Heavy Metal: Queen: Another One Bites the Dust – Heavy Metal: Van Halen: Jump . Music IR – Music? Music IR – Music? MOD MOD Similar to MIDI, but Some examples (from http://modarchive.org ) stores audio samples together with control instructions – Classical: Dark Castle (Part 1) should sound the same on every player – Classical: Canon in D – Classical: Beethoven: Für Elise a.k.a. tracker modules (first ever module creating program – Guitar: Sweet Lorraine was Soundtracker, created by Karsten Obarski 1987) – Latin: Heart and Soul – Techno: 10KBlur – Disco: Rob Hubbard . 3 Music IR – Music? Music IR – Music? Scores What is „Music“? Also referred to as „Sheet Music“ Music, of course! – Audio: wav, au, mp3, ... Hand-written or printed form of musical notation – Symbolic: MIDI, mod, ... – Handwritten scores www.samplesmith.com – Scores: Scan, MusicXML – Printed scores www.westminster.gov.uk – Typeset scores Text Community data Video/Images – MusicXML – Song lyrics – Playlists – Album covers Different IR tasks – Artis Biographies – Market basket – Music videos – Scan & Optical Music Recognition (OMR) – Websites: – Band evolution Fanpages, – Score following Album Reviews, – Melodic retrieval Genre descriptions . Music IR – Music? Music IR – Music? Handwritten scores Different styles of notation Handwritten / printed scores http://en.wikipedia.org/wiki/Musical_notation Different styles of notation – Neumes Ancient greek: – Staff stone at Delphi containing the second of the Complex annotations two hymns to Apollo Scanning scores Indian notation e.g. Musitek SmartScore: bhat notation http://www.musitek.com/ China Bach SheetmusicDemo: Quin notation http://bach.nau.edu/UWDigital/Washington.html . Music IR – Music? Music IR – Music? Music Typesetting / Scorewriter GNU LilyPond Software Software used to automate the task of writing and engraving sheet music, ako word processor for text http://lilypond.org/ Input via text editor or MIDI interface, Input: UTF-8, no graphical interface some support Scan+OMR some graphical editors produce LilyPond output Output: PS/PDF, graphics, MIDI, MusicXML (e.g.Rosegarden, NoteEdit, Canorus) Popular programs: – GNU LilyPond Software: http://lilypond.org/ Output: compiled to PDF, SVG, MIDI, ... – GUIDO Music Notation: http://www.salieri.org/GUIDO/ Notes are entered in note, pitch and length format – Finale: http://www.finalemusic.com/ – Sibelius: http://www.sibelius.com/ Used by several projects (Mutopia, Musipedia) – Comprehensive list: http://en.wikipedia.org/wiki/Scorewriter#Scorewriters . 4 Music IR – Music? Music IR – Music? LilyPond example LilyPond example (1/5, from http://en.wikipedia.org/wiki/GNU_LilyPond) (2/5, from http://en.wikipedia.org/wiki/GNU_LilyPond) #!lilypond firebreathers.ly -*- coding: utf-8; -*- %% Theme to "Fire Breathers", a homebrew NES game perpetually %% The header block defines the titles and texts. %% under development. Composed by Urpo Lankinen. %% Note: The composer has made this source code available \header { %% to Wikipedia under the GFDL license. Other versions outside title = "Theme to ``Fire Breathers!''" %% Wikipedia are typically under CC BY-SA license. instrument = "For the 2A03 or SID" %% This file uses Finnish note names (for example, where composer = "Urpo Lankinen" %% Americans use "F#" and "Bb", Finns use "Fis" and "B"). enteredby = "Urpo Lankinen" %% Dutch note names are used by default. updatedby = "Jan Nieuwenhuizen" \include "suomi.ly" date = "June 2005" %% Optional language upgrade helper. } \version "2.6.0" . Music IR – Music? Music IR – Music? LilyPond example LilyPond example (3/5, from http://en.wikipedia.org/wiki/GNU_LilyPond) (4/5, from http://en.wikipedia.org/wiki/GNU_LilyPond) Melody = \relative c'' { %% This is the second voice. \clef treble SecondVoice = \relative c { \time 3/4
Load more

Recommended publications
  • Improving Optical Music Recognition by Combining Outputs from Multiple Sources
    IMPROVING OPTICAL MUSIC RECOGNITION BY COMBINING OUTPUTS FROM MULTIPLE SOURCES Victor Padilla Alex McLean Alan Marsden Kia Ng Lancaster University University of Leeds Lancaster University University of Leeds victor.padilla. a.mclean@ a.marsden@ k.c.ng@ [email protected] leeds.ac.uk lancaster.ac.uk leeds.ac.uk ABSTRACT in the Lilypond format, claims to contain 1904 pieces though some of these also are not full pieces. The Current software for Optical Music Recognition (OMR) Musescore collection of scores in MusicXML gives no produces outputs with too many errors that render it an figures of its contents, but it is not clearly organised and unrealistic option for the production of a large corpus of cursory browsing shows that a significant proportion of symbolic music files. In this paper, we propose a system the material is not useful for musical scholarship. MIDI which applies image pre-processing techniques to scans data is available in larger quantities but usually of uncer- of scores and combines the outputs of different commer- tain provenance and reliability. cial OMR programs when applied to images of different The creation of accurate files in symbolic formats such scores of the same piece of music. As a result of this pro- as MusicXML [11] is time-consuming (though we have cedure, the combined output has around 50% fewer errors not been able to find any firm data on how time- when compared to the output of any one OMR program. consuming). One potential solution to this is to use Opti- Image pre-processing splits scores into separate move- cal Music Recognition (OMR) software to generate sym- ments and sections and removes ossia staves which con- bolic data such as MusicXML from score images.
    [Show full text]
  • Proceedings 2005
    LAC2005 Proceedings 3rd International Linux Audio Conference April 21 – 24, 2005 ZKM | Zentrum fur¨ Kunst und Medientechnologie Karlsruhe, Germany Published by ZKM | Zentrum fur¨ Kunst und Medientechnologie Karlsruhe, Germany April, 2005 All copyright remains with the authors www.zkm.de/lac/2005 Content Preface ............................................ ............................5 Staff ............................................... ............................6 Thursday, April 21, 2005 – Lecture Hall 11:45 AM Peter Brinkmann MidiKinesis – MIDI controllers for (almost) any purpose . ....................9 01:30 PM Victor Lazzarini Extensions to the Csound Language: from User-Defined to Plugin Opcodes and Beyond ............................. .....................13 02:15 PM Albert Gr¨af Q: A Functional Programming Language for Multimedia Applications .........21 03:00 PM St´ephane Letz, Dominique Fober and Yann Orlarey jackdmp: Jack server for multi-processor machines . ......................29 03:45 PM John ffitch On The Design of Csound5 ............................... .....................37 04:30 PM Pau Arum´ıand Xavier Amatriain CLAM, an Object Oriented Framework for Audio and Music . .............43 Friday, April 22, 2005 – Lecture Hall 11:00 AM Ivica Ico Bukvic “Made in Linux” – The Next Step .......................... ..................51 11:45 AM Christoph Eckert Linux Audio Usability Issues .......................... ........................57 01:30 PM Marije Baalman Updates of the WONDER software interface for using Wave Field Synthesis . 69 02:15 PM Georg B¨onn Development of a Composer’s Sketchbook ................. ....................73 Saturday, April 23, 2005 – Lecture Hall 11:00 AM J¨urgen Reuter SoundPaint – Painting Music ........................... ......................79 11:45 AM Michael Sch¨uepp, Rene Widtmann, Rolf “Day” Koch and Klaus Buchheim System design for audio record and playback with a computer using FireWire . 87 01:30 PM John ffitch and Tom Natt Recording all Output from a Student Radio Station .
    [Show full text]
  • Sibelius Artwork Guidelines Contents
    Sibelius Artwork Guidelines Contents Conditions of use ...........................................................................................................................3 Important information ..................................................................................................................4 Product names and logos.............................................................................................................5 Example copy..................................................................................................................................6 Endorsees ........................................................................................................................................7 Reviews............................................................................................................................................8 Awards...........................................................................................................................................11 House Style ...................................................................................................................................12 Conditions of use Who may use this material Authorized Sibelius distributors and dealers are permitted to reproduce text and graphics on this CD in order to market Sibelius products or PhotoScore, but only if these guidelines are adhered to, and all artwork is used unmodified and cleared by Sibelius Software before production of final proofs. Acknowledge trademarks Please
    [Show full text]
  • Musical Notation Codes Index
    Music Notation - www.music-notation.info - Copyright 1997-2019, Gerd Castan Musical notation codes Index xml ascii binary 1. MidiXML 1. PDF used as music notation 1. General information format 2. Apple GarageBand Format 2. MIDI (.band) 2. DARMS 3. QuickScore Elite file format 3. SMDL 3. GUIDO Music Notation (.qsd) Language 4. MPEG4-SMR 4. WAV audio file format (.wav) 4. abc 5. MNML - The Musical Notation 5. MP3 audio file format (.mp3) Markup Language 5. MusiXTeX, MusicTeX, MuTeX... 6. WMA audio file format (.wma) 6. MusicML 6. **kern (.krn) 7. MusicWrite file format (.mwk) 7. MHTML 7. **Hildegard 8. Overture file format (.ove) 8. MML: Music Markup Language 8. **koto 9. ScoreWriter file format (.scw) 9. Theta: Tonal Harmony 9. **bol Exploration and Tutorial Assistent 10. Copyist file format (.CP6 and 10. Musedata format (.md) .CP4) 10. ScoreML 11. LilyPond 11. Rich MIDI Tablature format - 11. JScoreML RMTF 12. Philip's Music Writer (PMW) 12. eXtensible Score Language 12. Creative Music File Format (XScore) 13. TexTab 13. Sibelius Plugin Interface 13. MusiXML: My own format 14. Mup music publication program 14. Finale Plugin Interface 14. MusicXML (.mxl, .xml) 15. NoteEdit 15. Internal format of Finale (.mus) 15. MusiqueXML 16. Liszt: The SharpEye OMR 16. XMF - eXtensible Music 16. GUIDO XML engine output file format Format 17. WEDELMUSIC 17. Drum Tab 17. NIFF 18. ChordML 18. Enigma Transportable Format 18. Internal format of Capella (ETF) (.cap) 19. ChordQL 19. CMN: Common Music 19. SASL: Simple Audio Score 20. NeumesXML Notation Language 21. MEI 20. OMNL: Open Music Notation 20.
    [Show full text]
  • Notensatz Mit Freier Software
    Notensatz mit Freier Software Edgar ’Fast Edi’ Hoffmann Community FreieSoftwareOG [email protected] 30. Juli 2017 Notensatz bezeichnet (analog zum Textsatz im Buchdruck) die Aufbereitung von Noten in veröffentlichungs- und vervielfältigungsfähiger Form. Der handwerkliche Notensatz durch ausgebildete Notenstecher bzw. Notensetzer wird seit dem Ende des 20. Jahrhunderts vom Computernotensatz verdrängt, der sowohl bei der Druckvorlagenherstellung als auch zur Verbreitung von Musik über elektronische Medien Verwendung findet. Bis in die zweite Hälfte des 15. Jahrhunderts konnten Noten ausschließlich handschriftlich vervielfältigt und verbreitet werden. Notensatz Was bedeutet das eigentlich? 2 / 20 Der handwerkliche Notensatz durch ausgebildete Notenstecher bzw. Notensetzer wird seit dem Ende des 20. Jahrhunderts vom Computernotensatz verdrängt, der sowohl bei der Druckvorlagenherstellung als auch zur Verbreitung von Musik über elektronische Medien Verwendung findet. Bis in die zweite Hälfte des 15. Jahrhunderts konnten Noten ausschließlich handschriftlich vervielfältigt und verbreitet werden. Notensatz Was bedeutet das eigentlich? Notensatz bezeichnet (analog zum Textsatz im Buchdruck) die Aufbereitung von Noten in veröffentlichungs- und vervielfältigungsfähiger Form. 2 / 20 Bis in die zweite Hälfte des 15. Jahrhunderts konnten Noten ausschließlich handschriftlich vervielfältigt und verbreitet werden. Notensatz Was bedeutet das eigentlich? Notensatz bezeichnet (analog zum Textsatz im Buchdruck) die Aufbereitung von Noten in veröffentlichungs-
    [Show full text]
  • Improvisatory Music and Painting Interface
    Improvisatory Music and Painting Interface Hugo Solís García Licenciado en Piano Universidad Nacional Autónoma de México November 2001 Submitted to the Program in Media Arts and Sciences, School of Architecture and Planning, in partial fulfillment of the requirements for the degree of Master of Science in Media Arts and Science at the Massachusetts Institute of Technology September 2004 © 2004 Massachusetts Institute of Technology All rights reserved Author: Hugo Solís García Program in Media Arts and Sciences August 16, 2004 Certified by: Tod Machover Professor of Music and Media Thesis Supervisor, MIT Program in Media Arts and Sciences Accepted by: Dr. Andrew B. Lippman Chair, Departmental Committee on Graduate Students Program in Media Arts and Sciences Title Improvisatory Music and Painting Interface Hugo Solís García Submitted to the Program in Media Arts and Sciences, School of Architecture and Planning on August 16, 2004. in partial fulfillment of the requirements for the degree of Master Of Science in Media Arts and Sciences Massachusetts Institute of Technology Abstract Shaping collective free improvisations in order to obtain solid and succinct works with surprising and synchronized events is not an easy task. This thesis is a proposal towards that goal. It presents the theoretical, philosophical and technical framework of the Improvisatory Music and Painting Interface (IMPI) system: a new computer program for the creation of audiovisual improvisations performed in real time by ensembles of acoustic musicians. The coordination of these improvisations is obtained using a graphical language. This language is employed by one “conductor” in order to generate musical scores and abstract visual animations in real time.
    [Show full text]
  • Using Smartscore 2.Pdf
    Using SmartScore Scanning Music Be sure you have the necessary scanner drivers installed before attempting to scan from inside SmartScore. Most scanners come with software that enable programs such as SmartScore to control them. TWAIN drivers and/or Mac plug-ins are normally included in the software packaged with most scanners. It may be necessary for certain Mac users to perform a “Custom > TWAIN” installation from the CD accompanying your scan- ner; depending on the manufacturer. NOTE: Scanner drivers are often updated by scanner manufacturers and posted on their web sites. If problems occur during scanning, it is always a good idea to check the Internet for updated scanner drivers before calling Musitek Technical Support. Mac Users: Skip the next section. Turn to “Scanning in Macintosh” on page 6. Scanning in Windows: Using the SmartScore Scanning Interface a. Push the Scan button in the Navigator or in the Main Toolbar. Figure 1: Scan Button b. If there is no response, go to File > Scan Music > Select Scanner and choose appropriate TWAIN driver. If you do not see anything listed in the Select Scanner window, your drivers are probably not installed. Install or replace TWAIN driver from scanner CD or from “Driver Download” area of scanner manufacturer’s website. c. If the scanner still does not operate properly, go to “Choosing an alternative scanning interface” on page 5. USING SmartScore 1 Help > Using SmartScore Your scanner should immediately begin to operate with Scan or Acquire. A low-resolution pre-scan should soon appear in the Preview window. FIGURE 2: SmartScore scanning interface d.
    [Show full text]
  • Lilypond Informations Générales
    LilyPond Le syst`eme de notation musicale Informations g´en´erales Equipe´ de d´eveloppement de LilyPond Copyright ⃝c 2009–2020 par les auteurs. This file documents the LilyPond website. Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.1 or any later version published by the Free Software Foundation; with no Invariant Sections. A copy of the license is included in the section entitled “GNU Free Documentation License”. Pour LilyPond version 2.21.82 1 LilyPond ... la notation musicale pour tous LilyPond est un logiciel de gravure musicale, destin´e`aproduire des partitions de qualit´e optimale. Ce projet apporte `al’´edition musicale informatis´ee l’esth´etique typographique de la gravure traditionnelle. LilyPond est un logiciel libre rattach´eau projet GNU (https://gnu. org). Plus sur LilyPond dans notre [Introduction], page 3, ! La beaut´epar l’exemple LilyPond est un outil `ala fois puissant et flexible qui se charge de graver toutes sortes de partitions, qu’il s’agisse de musique classique (comme cet exemple de J.S. Bach), notation complexe, musique ancienne, musique moderne, tablature, musique vocale, feuille de chant, applications p´edagogiques, grands projets, sortie personnalis´ee ainsi que des diagrammes de Schenker. Venez puiser l’inspiration dans notre galerie [Exemples], page 6, 2 Actualit´es ⟨undefined⟩ [News], page ⟨undefined⟩, ⟨undefined⟩ [News], page ⟨undefined⟩, ⟨undefined⟩ [News], page ⟨undefined⟩, [Actualit´es], page 103, i Table des mati`eres
    [Show full text]
  • Some Experience with Musixtex
    Some Experience with MusiXTEX Jean-Michel HUFFLEN LIFC | University of Franche-Comt´e BachoTEX, 29 April 2011 1 Contents Who am I? What is MusiXTEX? LATEX, ConTEXt, etc. Difficult typography A musician's point of view Conclusion 2 Musical CV In parallel with `classical' studies (Mathematics, Computer Science): Musical CV In parallel with `classical' studies (Mathematics, Computer Science): School of Music. Musical CV In parallel with `classical' studies (Mathematics, Computer Science): School of Music. High diploma in music training, bassoon, harmony, counterpoint (1981). Musical CV In parallel with `classical' studies (Mathematics, Computer Science): School of Music. High diploma in music training, bassoon, harmony, counterpoint (1981). I played in various orchestras (sometimes as a conductor). 3 As a composer Not very attracted by vocal music. As a composer Not very attracted by vocal music. Symphonies, concertos, chamber music pieces. As a composer Not very attracted by vocal music. Symphonies, concertos, chamber music pieces. Atonal style, then synthesis between tonal- ity/modality and atonality. 4 What is MusiXTEX Aims to typeset high-quality print output scores. What is MusiXTEX Aims to typeset high-quality print output scores. Authors warn: Let us recall that TEX was not designed for scores, but for texts. What is MusiXTEX Aims to typeset high-quality print output scores. Authors warn: Let us recall that TEX was not designed for scores, but for texts. Stalled, since Daniel Taupin's death. 5 Three-step system • [pdf]latex filename • musixflx filename • [pdf]latex filename 6 LATEX, ConTEXt, etc. Aim to get high-quality print outputs. LATEX, ConTEXt, etc. Aim to get high-quality print outputs.
    [Show full text]
  • MIDI Keyboard MIDI Keyboard
    2020/08/11 08:44 1/10 MIDI Keyboard MIDI Keyboard Tento návod popisuje zprovoznění a používání (nejen) MIDI keyboardu v Ubuntu. Instalace Návod byl testován na Ubuntu 14.04 Trusty Tahr a 15.04 Vivid Vervet a keyboardu FOX KeyControl 49. Nejdříve ze všeho musíme mít něco, co vydává zvuk, tedy syntetizér + přepojovač. QSynth instaluje soundfont GM, který je větší a má více instrumentů. Pokud šetříte místo na disku, nainstalujte soundfont GS: fluid-soundfont-gs, potom GM odstraňte a v nastavení QSynth nastavte GS. Nainstalujeme tedy balíky: QSynth (Syntetizér), QjackCtl (přepojovač) Celkově těmto dvěma aplikacím budeme říkat JACK Server Spuštění Nejdříve musíme spustit QSynth. Až se okénko kompletně načte, klikneme vlevo na tlačítko Nastavení…, zvolíme záložku Zvuková písma → Otevřít… a zvolíme si soundfont. Nyní už můžeme připojit náš MIDI Keyboard. Potom zapneme QjackCtl (ale QSynth s ním necháme běžet) a k tomu zapneme nějakou aplikaci, kterou chcete keyboardem ovládat. Přepneme se zpět do QjackCtl, kde klikneme na tlačítko Spojit. Tam (nejčastěji v záložce ALSA-MIDI) spojíme výstup Midi Keyboard (neplést s Midi Trough) se vstupem aplikace, kterou chceme ovládat a tu spojíme s výstupem FLUID Synth. I po vypnutí Qsynth zůstane na pozadí běžet JACK daemon, který zastavuje zvuk. Ten se musí Ubuntu CZ/SK - https://wiki.ubuntu.cz/ Last update: 2019/02/25 18:21 midi_keyboard https://wiki.ubuntu.cz/midi_keyboard vypnout přes Správce úloh (v Ubuntu Sledování systému) → zabít proces jackdbus. Aplikace Několik aplikací pro použití MIDI keyboardu, všechny uvedené jsou zdarma. Přehrávání Pro pouhé hraní v reálném čase stačí v QjackCtl spojit výstup Midi Keyboard se vstupem Fluid Synth.
    [Show full text]
  • Improved Optical Music Recognition (OMR)
    Improved Optical Music Recognition (OMR) Justin Greet Stanford University [email protected] Abstract same index in S. Observe that the order of the notes is implicitly mea- This project focuses on identifying and ordering the sured. Any note missing from the output, out of place, or notes and rests in a given measure of sheet music using a erroneously added heavily penalizes the percentage. More novel approach involving a Convolutional Neural Network qualitatively, we can render the output of the algorithm as (CNN). Past efforts in the field of Optical Music Recogni- sheet music and visually check how it compares to the input tion (OMR) have not taken advantage of neural networks. measure. The best architecture we developed involves feeding pro- Multiple commercial OMR products exist and their ef- posals of regions containing notes or rests to a CNN then fectiveness has been studied to be imperfect enough to pre- removing conflicts in regions identifying the same symbol. clude many practical applications [4, 20]. We run the test We compare our results with a commercial OMR product images through one of them (SmartScore [12]) and use the and achieve similar results. evaluation criteria described above to serve as a benchmark for the proposed algorithm. 1. Introduction 2. Related Work Optical Music Recognition (OMR) is the problem of The topic of OMR is heavily researched. The research converting a scanned image of sheet music into a symbolic can be broadly broken up into three categories: evaluation representation like MusicXML [9] or MIDI. There are many of existing methods, discovery of novel solutions for spe- obvious practical applications for such a solution, such as cific parts of OMR, and classical approaches to OMR.
    [Show full text]
  • Efficient Optical Music Recognition Validation Using MIDI Sequence Data by Janelle C
    Efficient Optical Music Recognition Validation using MIDI Sequence Data by Janelle C. Sands Submitted to the Department of Electrical Engineering and Computer Science in partial fulfillment of the requirements for the degree of Master of Engineering in Electrical Engineering and Computer Science at the MASSACHUSETTS INSTITUTE OF TECHNOLOGY May 2020 ○c Massachusetts Institute of Technology 2020. All rights reserved. Author................................................................ Department of Electrical Engineering and Computer Science May 12, 2020 Certified by. Michael Scott Cuthbert Associate Professor Thesis Supervisor Accepted by . Katrina LaCurts Chair, Master of Engineering Thesis Committee ii Efficient Optical Music Recognition Validation using MIDI Sequence Data by Janelle C. Sands Submitted to the Department of Electrical Engineering and Computer Science on May 12, 2020, in partial fulfillment of the requirements for the degree of Master of Engineering in Electrical Engineering and Computer Science Abstract Despite advances in optical music recognition (OMR), resultant scores are rarely error-free. The power of these OMR systems to automatically generate searchable and editable digital representations of physical sheet music is lost in the tedious manual effort required to pinpoint and correct these errors post-OMR, or evento just confirm no errors exist. To streamline post-OMR error correction, I developeda corrector to automatically identify discrepancies between resultant OMR scores and corresponding Musical Instrument Digital Interface (MIDI) scores and then either automatically fix errors, or in ambiguous cases, notify the user to manually fix errors. This tool will be open source, so anyone can contribute to further improving the accuracy of OMR tools and expanding the amount of trusted digitized music. Thesis Supervisor: Michael Scott Cuthbert Title: Associate Professor iii iv Acknowledgments Thank you to my advisor, Professor Cuthbert, for sharing your enthusiasm, expertise, and encouragement throughout this project.
    [Show full text]