FINAL PROGRAM

AUGUST 18–20, 2008 NEW YORK MARRIOTT MARQUIS NEW YORK CITY

SPEECH IN THE MAINSTREAM

WWW. SPEECHTEK. COM

KEYNOTE SPEAKERS

w Analytics w Mobile Devices Ray Kurzweil author of The Age of Spiritual Machines and w Natural Language w Multimodal The Age of Intelligent Machines w Security w Video and Speech w Voice User Interfaces w Testing and Tuning Lior Arussy President, Strativity Group author of Excellence Every Day

Gold Sponsors:

Kevin Mitnick author, world-famous (former) hacker, & Silver Sponsor: Media Sponsors: security consultant Organized and produced by Welcome to SpeechTEK 2008 Speech in the Mainstream

Last year we recognized that the speech industry was at a tipping Conference Chairs point, or the point at which change becomes unstoppable. To James A. Larson illustrate this point, we featured Malcolm Gladwell, author of the VP, Larson Technical Services best-selling book, The Tipping Point, as our keynote speaker. Not surprisingly, a lot has happened since the industry tipped: We’re Susan L. Hura seeing speech technology in economy-class cars and advertised on television to millions of households, popular video games with VP, Product Support Solutions more robust voice commands, and retail shelves stocked with affordable, speech-enabled aftermarket navigation systems. It’s Director, SpeechTEK clear that speech is in the mainstream—the focus of this year’s Program Planning conference. David Myron, Editorial Director, Speech technology has become mainstream for organizations CRM and Speech Technology magazines seeking operational efficiencies through self-service. For enterprises, speech enables employees to reset passwords, sign up for benefits, and find information on company policies and procedures. In healthcare, doctors have long used speech dictation systems to record symptoms and diagnosis; healthcare organizations are beginning to use speech to contact patients for appointment management, report test results, and for healthcare management. Government organizations use speech technologies to better serve their constituencies. During the 2007 California wildfire season, local government voice systems notified thousands of homeowners of potential fire danger, which was likely the leading reason for so few deaths. Speech has clearly made a positive impact on many people and organizations. At SpeechTEK 2008, a variety of organizations will explain how they are succeeding with speech technology as well. SpeechTEK 2008 is the world’s biggest speech technology conference and exhibition. Nowhere else can you find so much information on the latest ideas, technologies, services, and solutions. This year, we offer more leading- edge information and opportunities for interaction with industry movers and shakers. Keep your company in the speech technology mainstream by attending SpeechTEK 2008 and discussing how these exciting new technologies will drive the future. This program describes the many activities and opportunities that will help your company chart a successful course with speech technology. Welcome to SpeechTEK 2008, our best conference ever. Please recycle this program.

HEAR THESE TOPICS AND MORE: PLUS SPECIAL TRACKS: Meeting Business Goals w The state of the speech w How to monitor and manage your with Speech Technology technology industry speech application throughout its w A new track designed specifically for lifecycle w What you need to know to Business Executives and Managers get started on a speech project w Techniques for using nontraditional Advanced Speech data to deliver a rapid round of tuning w Getting management’s Technologies Symposium commitment early in the process w How human guides have improved w A 2-day track covering advanced service for retail self-service dialogue techniques, machine translation, natural language, and more. applications

2 AUGUST 18–20, 2008 w NEW YORK MARRIOTT MARQUIS, NEW YORK CITY CONFERENCE OVERVIEW

Evening Receptions SpeechTEK invites attendees with a Networking Pass or above Table of Contents to a Welcome Reception Monday evening in the Exhibit Hall from 5:00 p.m. – 7:00 p.m. sponsored by Angel.com as well as the Tuesday evening reception from 5:30 p.m. – 7:00 p.m. on the 9th floor. Mingle with exhibitors, speakers, and conference 4 w Keynotes attendees while enjoying a glass of wine and hors d’oeuvres. 6 w Conference At-A-Glance CD-ROMs Most sessions are being recorded and CD-ROMs may be pur- 8 w SpeechTEK University chased at the conference. Visit the Digital Record table (please ask at the registration desk for exact location) for more infor- 10 w Monday Conference mation or visit www.digitalrecord.org. 18 w Tuesday Conference Press Room/Media Center sponsored by Wednesday Conference The press area will be available in the Gilbert Room on the 4th 26 w floor Monday from 9:00 a.m. – 7:00 p.m., Tuesday from 9:00 a.m. – 6:30 p.m., and Wednesday from 9:00 a.m. – 4:00 p.m. 32 w SpeechTEK Labs Speaker Directory Presentation Links 34 w Get access to many PowerPoint presentations and websites Exhibitor List & Floor Plan used in conjunction with the SpeechTEK 2008 sessions! Links 40 w will be posted on the SpeechTEK.com website soon after the Exhibitor Directory conference. 42 w 46 Company Workshops Sunrise Semester Discussions w SpeechTEK will host early morning discussions for business professionals, project leaders, designers, and developers to openly talk about a variety of topics. This is a great opportu- Continental Breakfast & Coffee Breaks nity for attendees to network, discuss challenges, and share A continental breakfast will be provided for conference atten- ideas with peers and speech technology professionals. dees each morning from 8:00 a.m. to 9:00 a.m. before the keynote session. Additional coffee breaks will take place each morning and afternoon. Please check the schedule for exact SpeechTEK University times and locations. If you are considering deploying a speech application or look- ing to increase your knowledgebase, be sure to attend a Speech Technology Magazine Awards SpeechTEK University course. SpeechTEK University courses While at the conference, find out the winners of Speech Tech- are in-depth, focused, 3-hour seminars on topics of special nology magazine’s 2008 Speech Industry awards. interest to speech technology and information technology pro- fessionals taught by experienced instructors. These are sepa- rately priced or may be purchased as part of your conference Email Stations registration. See pages 8 – 9 for more details. Email stations will be open in the Exhibit Hall during regular exhibit hours for attendees to check their e-mail. SpeechTEK Labs Exhibit Hall Hours This year marks our first SpeechTEK Labs, in which attendees Monday, August 18...... 5:00 p.m. – 7:00 p.m. can sample the latest speech technology products, talk to developers, and speak openly to our expert panel of judges Tuesday, August 19...... 10:00 a.m. – 5:30 p.m. about the technology. These sessions, which take place in the Wednesday, August 20...... 10:00 a.m. – 2:00 p.m. Exhibit Hall on Tuesday and Wednesday, will provide objective evaluations of new speech technology and show how speech Registration Desk Hours is breaking new ground in various industries. Attendees will Sunday, August 17 ...... 1:00 p.m. – 7:00 p.m. have the opportunity to evaluate speech products in three cat- Monday, August 18 ...... 7:00 a.m. – 7:00 p.m. egories: multimodal applications, speech tools, and voice-acti- Tuesday, August 19 ...... 7:30 a.m. – 5:30 p.m. vated games. As an added benefit, after the lab, the judges will issue ratings of the tools to provide further guidance. For Wednesday, August 20 ...... 8:00 a.m. – 2:00 p.m. complete descriptions, see page 32. Thursday, August 21 ...... 8:00 a.m. – 1:30 p.m.

WWW.SPEECHTEK.COM 3 KEYNOTES

All Keynotes and Sponsored Keynote lunches will take place in the Broadway Ballroom on the 6th floor.

Preparing for the New Consumer TUESDAY w 9:00 a.m. – 10:00 a.m. Lior Arussy, author of several books, including his most recent, Excellence Every Day: Make the Speech in the Mainstream and Beyond Daily Choice—Inspire Your Employees and MONDAY w 9:00 a.m. – 10:00 a.m. Amaze Your Customers (Information Today, Inc.) Ray Kurzweil, author of The Age of Spiritual Despite technological innovations, companies Machines: When Computers Exceed Human are failing to meet increasing customer expectations. To stay Intelligence and The Age of Intelligent Machines competitive, organizations must meet customers’ growing We’re living in a world of rapid technological inno- demands by incorporating new ways of thinking, connecting, vation and increasing pervasiveness of informa- and doing business that will create profitable and delightful cus- tion technology. Consumers not only appreciate these advances tomer experiences. Lior Arussy is a renowned author; business in their professional and personal lives, they are demanding it. visionary; creative catalyst; and president of Strativity Group, a What does this mean for those who are driving development consultancy which advises Global 2000 companies and emerg- and acceptance of speech technology? Glean insight from one ing businesses around the world. Arussy will address the biggest of the most distinguished speech technology innovators of our customer strategy issues facing organizations today, such as the time, Ray Kurzweil, who will share his views on how speech tech- new rules of customer engagement, and offer suggestions on nology will work in conjunction with other emerging technolo- how to better prepare for the new consumer. gies to bring us to an age of intelligent machines. The Art of Deception: SPONSORED KEYNOTE LUNCHEONS Are You in Danger of Being ‘Conned’? WEDNESDAY w 8:45 a.m. – 10:00 a.m. Phone 2.0: The Mobile Kevin Mitnick, author, world-famous (former) Generation at Work Sponsored by hacker, and security consultant MONDAY w 12:00 p.m. – 1:15 p.m. As more businesses and consumers depend on Beatriz V. Infante, President and CEO, speech technology, it’s becoming increasingly VoiceObjects, Inc. important to protect your investments. With more than 15 years of experience in exploring computer security, Kevin Mitnick is a largely self-taught expert in exposing the vulnerabil- Remember Me: A Case Study ities of complex operating systems and telecommunications on Personalized Self-Service Sponsored by devices. As the world’s most famous (former) hacker, Mitnick has TUESDAY w 12:30 p.m. – 1:45 p.m. been the subject of countless news and magazine articles pub- Tracy Griffith, Sr. Manager Customer Technology Reservations lished throughout the world. Mitnick will share how hackers com- & Premium Services, American Airlines promise your security and what you can do about it.

SpeechTEK Labs TUESDAY, AUGUST 19 WEDNESDAY, AUGUST 20 10:45 a.m. – 12:30 p.m. 10:45 a.m. – 12:30 p.m. This year marks our first SpeechTEK Labs, in which atten- Tools for Speech Applications Multimodal Application Lab dees can sample the latest speech technology products, MODERATOR: MODERATOR: talk to developers, and speak openly to our expert panel of Moshe Yudkowsky, Deborah Dahl, Principal, judges about the technology. These sessions, which take Disaggregate Conversational Technologies place in the Exhibit Hall on Tuesday and Wednesday, will provide objective evaluations of new speech technology 1:45 p.m. – 3:30 p.m. and show how speech is breaking new ground in various Voice-Activated Games industries. Attendees will have the opportunity to evaluate For complete speech products in three categories: multimodal applica- MODERATOR: tions, speech tools, and voice-activated games. As an David Thomson, CTO, descriptions and added benefit, after the lab, the judges will issue ratings of SpeechPhone company listings, the tools to provide further guidance. see page 32.

4 AUGUST 18–20, 2008 w NEW YORK MARRIOTT MARQUIS, NEW YORK CITY AD ALL KEYNOTES WILL BE HELD IN THE BROADWAY BALLROOM, 6TH FLOOR. 9:00 a.m. – 10:00 a.m. WELCOME & OPENING KEYNOTE w Speech in the Mainstream and Beyond w Ray Kurzweil TRACK A w Soho Complex (7th Floor) TRACK B w Empire Complex (7th Floor) MEETING BUSINESS GOALS WITH SPEECH TECHNOLOGY VUI DESIGN PRINCIPLES & TECHNIQUES 1 10:15 a.m. – 1 Mainstream Speech 2008: The State of the Industry What End Users Really Want 0 0 1 11:00 a.m. 1 Tim Pearce, Mike Bergelson, Bonnie Crater, Monique Bozeman Jenni McKienzie, Moshe Yudkowsky B A 2 11:15 a.m. – 2 Planning for Speech Managing Client/Vendor Relationships 0 0 1 12:00 p.m. 1 Catherine Zhu, Caroline Nelson David Martin, Jessica Stevens & Jenny Burr B A 12:00 p.m. – 1:15 p.m. SPONSORED KEYNOTE LUNCH w Sponsored by w Phone 2.0: The Mobile Generation at Work w Beatriz V. Infante 3 1:15 p.m. – 3 Preventive Medicine: Keeping Your Speech System Healthy Reverse Navigation Hands-On Session 0 0 1 2:00 p.m. 1 Fran McTernan, Ingo Bors Peter Krogh B A 4 2:15 p.m. – 4 Agile Speech Projects Universal Commands 0 0 1 3:00 p.m. 1 Charles Lewis, Leslie Carroll Walker James Mesbur & Karen Moylé, Rita Dhruve B A 5 3:15 p.m. – CCS: Speech in Financial Services 5 Lessons in Multimodal Usability 0 0 1 4:00 p.m. Glen Graham, Ricardo de Carvalho Destro 1 Eduardo Olvera, Matt Prather A B 6 4:15 p.m. – 6 The Business of Speech Analytics How May I Help You? 0 0 1 5:00 p.m. 1 Michael Codini, Greg Borton, & Cliff LaCoursiere Charles Galles, Mary Constance Parks B A 5:00 p.m. – 7:00 p.m. EXHIBIT HALL GRAND OPENING & WELCOME RECEPTION w In the Exhibit Hall w Sponsored by Angel.com

9:00 a.m. – 10:00 a.m. KEYNOTE w Preparing for the New Consumer w Lior Arussy 10:00 a.m. – 10:45 a.m. BREAK w Visit the Exhibits TRACK A w Soho Complex (7th Floor) TRACK B w Empire Complex (7th Floor) MEETING BUSINESS GOALS WITH SPEECH TECHNOLOGY VUI DESIGN PRINCIPLES & TECHNIQUES 1 10:45 a.m. – 1 Speech: Thinking Out of the Box How Do You VUI? 0 0 2 11:30 a.m. 2 Moshe Yudkowsky, Whitney Quesenberry, Steve Chirokas Bruce Balentine & Ken Rehor, David Suenderman & Ethan Levine B A 2 11:45 a.m. – 2 Video in the Network and Call Center 0 0 Inaugural Meeting of the Association for Voice Interaction Design 2 12:30 p.m. 2 Bob Cooper, Rob Marchand B A 12:30 p.m. – 1:45 p.m. SPONSORED KEYNOTE LUNCH w Sponsored by w A Case Study on Personalized Self-Service w Tracy Griffith 3 1:45 p.m. – 3 Cross-Channel User Experience Business Problems, VUI Answers 0 0 2 2:30 p.m. 2 Lizanne Kaiser, Elaine Cascio Mark Webb, Daniel Padgett B A 4 2:45 p.m. – 4 Speech in CPG/Retail VUI Design for Multimodal Applications 0 0 2 3:30 p.m. 2 Tom Scott, Michael Jacks David Thornton, Juan Gilbert B A 3:30 p.m. – 4:15 p.m. BREAK w Visit the Exhibits 5 4:15 p.m. – 5 The VUI Backlash Driving the Conversation Forward in the Face of Errors 0 0 2 5:15 p.m. 2 Ahmed Bouzid, Simonie Wilson, Nick Ezzo Simsar; Bloom & Liscombe; Peterson B A 5:30 p.m. – 7:00 p.m. RECEPTION w On the 9th Floor

8:45 a.m. – 10:00 a.m. KEYNOTE w The Art of Deception: Are You in Danger of Being ‘Conned’? w Kevin Mitnick 10:00 a.m. – BREAK Visit the Exhibits 10:45 a.m. w TRACK A w Soho Complex (7th Floor) TRACK B w Empire Complex (7th Floor) MEETING BUSINESS GOALS WITH SPEECH TECHNOLOGY VUI DESIGN PRINCIPLES & TECHNIQUES 1 10:45 a.m. – 1 Customer Satisfaction 2008 Making VUIs Even Better 0 0 3 11:30 a.m. 3 Lynda Kate Smith, Michael Codini Patrick Nguyen, Bob Cooper B A 2 11:45 a.m. – 2 DTMF Versus Speech Designing for Spanish 0 0 3 12:30 p.m. 3 Darla Tucker, Matt Whipple Sondra Ahlen, Jose Elizondo B A 12:30 p.m. – 1:45 p.m. LUNCH BREAK w Visit the Exhibits 3 1:45 p.m. – 3 Security for Speech Happy Together? Speech & Touchtone 0 0 3 2:30 p.m. 3 Steven Cawn, Kris Herrin David Attwater, Phil Shinn B A 4 2:45 p.m. – 4 The Crystal Ball The Ultimate VUI Showdown 0 0 3 3:30 p.m. 3 Judith Markowitz, Daniel Hong, Bill Meisel Balentine, Bouzid, Brill, Kaiser, Krogh & Strand B A

Sunrise Semester Discussions w 8:00 a.m. – 8:50 a.m. MONDAY, AUGUST 18

SpeechTEK will host early morning discussions in the Royale Room DISCUSSION 1 Emotion in Speech Technology w Baggia (6th floor) for business professionals, project leaders, designers, and DISCUSSION 2 The Effectiveness of Marketing Messages w Rogers developers to openly talk about a variety of topics.

6 AUGUST 18–20, 2008 w NEW YORK MARRIOTT MARQUIS, NEW YORK CITY CONFERENCE AT-A-GLANCE

MONDAY, AUGUST 18 TRACK C w Majestic Complex (6th Floor) TRACK D w Shubert Complex (6th Floor) ADVANCED SPEECH TECHNOLOGIES SYMPOSIUM DEVELOPMENT AND DEPLOYMENT 1 1 Advanced ASR: A Global Perspective Design Methodologies and Tools 0 0 1 1 Roberto Pieraccini Matthew Yuschik, Leah Eyler D C 2 2 Application of Advanced ASR Speaking and Listening to Mobile Devices 0 0 1 1 Marsal Gavalda, Vlad Sejnoha Matthew Yuschik D C

SPONSORED KEYNOTE LUNCH w Sponsored by w Phone 2.0: The Mobile Generation at Work w Beatriz V. Infante 3 3 Success With Machine Translation New Standard Languages for Developing Speech Applications 0 0 1 1 Hannah Grap, Imed Zitouni Paolo Baggia, Jim Barnett D C 4 4 Advanced Dialogue’s Growing Sophistication Advanced Techniques for Using Grammars 0 0 1 1 Phillip Hunter, David Thomson Dominique Boucher, David Claiborn C D 5 5 Natural Language Processing Techniques Integrating Speech Technologies With Enterprise Applications 0 0 1 1 Peter Trompetter, Daniel O’Sullivan Michael Codini, John Oh D C 6 6 IVR Virtualization The Impact of W3C Standard Languages 0 0 1 1 Rob Kassel, Nava Shaked R.J. Auburn, Ken Rehor C D EXHIBIT HALL GRAND OPENING & WELCOME RECEPTION w In the Exhibit Hall w Sponsored by Angel.com TUESDAY, AUGUST 19 BREAK w Visit the Exhibits TRACK C TRACK C w Majestic Complex (6th Floor) TRACK D w Shubert Complex (6th Floor) ADVANCED SPEECH TECHNOLOGIES SYMPOSIUM DEVELOPMENT AND DEPLOYMENT 1 1 Natural Language Processing: Automaticity Securing CCXML and VoiceXML Applications 0 0 2 2 Mark Pfeiffer, Emmett Coin Dan York D C 2 2 Enhanced VUI: Context-Sensitive Dialogue Personalization and Context 0 0 2 2 Sunil Issar, Susan Boyce Patrick Nguyen, Steve Pollock D C SPONSORED KEYNOTE LUNCH w Sponsored by w A Case Study on Personalized Self-Service w Tracy Griffith 3 3 Enhanced VUI: Dynamic Response Generation CCS: Speech in Healthcare 0 0 2 2 Ahmed Bouzid, Phil Shinn Val Nenov, David Engelhardt D C 4 4 Advances in Automotive Speech CCS: Speech in Government/Utilities 0 0 2 2 Roberto Sicconi, Tom Schalk Marge Vizcarra & James Whitten, Michael Roberts C D BREAK w Visit the Exhibits 5 5 Ancillary Transcription Techniques CCS: Speech in Telecom 0 0 2 2 Michael Picheny, Jordan Cohen Manohar Kesireddy, Fred MacKenzie, Charles Giordano C D RECEPTION w On the 9th Floor WEDNESDAY, AUGUST 20 BREAK w Visit the Exhibits TRACK C w Majestic Complex (6th Floor) TRACK D w Shubert Complex (6th Floor) TODAY’S SPEECH TECHNOLOGIES DEVELOPMENT AND DEPLOYMENT 1 1 Video and Speech CCS: Speech in Financial Services/Insurance 0 0 3 3 Andrew Fuller, Valentine Matula Brent Sparks & Remus Siclovan D C 2 2 CCS: Speech in Automotive Testing 0 0 3 3 Ben Miners, Christian McCarrick Jim Jenkins; Judi Halperin & Dennis Jakobsen D C LUNCH BREAK w Visit the Exhibits 3 3 Multimodal Standards and Applications CCS: Speech in Customer Service 0 0 3 3 Deborah Dahl & Ingmar Kliche; Raj Tumuluri Richard Grant, Chris Weeks D C 4 4 Developing and Testing Multimodal Applications Tuning Tips 0 0 3 3 David Pelland, Cliff McEnery Yves Normandin, Stephen Keller C D

TUESDAY, AUGUST 19 WEDNESDAY, AUGUST 20 DISCUSSION 3 Speaker Identification and Verification w Markowitz DISCUSSION 6 Speech Development Teams w Shaked DISCUSSION 4 It Ain’t Shakespeare…Or Is It? w Auckland & Polkosky DISCUSSION 7 Project & Perfect w Tranter & Turney DISCUSSION 5 W3C Multimodal Working Group w Dahl DISCUSSION 8 Analytics w Brinsmead WWW.SPEECHTEK.COM 7 SpeechTEK University courses provide At-A-Glance in-depth, 3-hour seminars on compelling

y topics for speech technology and information t i s r technology professionals. Experienced e SUNDAY, AUGUST 17 v i 1:30 p.m. – 4:30 p.m. n speech technology practitioners teach each U STKU-1

K class in an intimate classroom setting to

E Introduction to Voice User

T foster an educational and structured learning

h Interface Design c Eduardo Olvera e experience. If you are considering deploying e p

S STKU-2 a speech application, looking to increase your Introduction to Speech Technologies James A. Larson knowledgebase in one of the key areas below, or you simply need a speech THURSDAY, AUGUST 21 9:00 a.m. – 12:00 p.m. technology refresher, attend a SpeechTEK STKU-3 University course. These courses are Introduction to Grammar Writing separately priced or may be purchased as Jenni McKienzie & Judi Halperin part of your conference registration. STKU-4 Guerilla Tactics in VUI Design Melanie Polkosky SUNDAY, AUGUST 17 STKU-5 Natural Language Introduction to Design STKU-1 Deborah Dahl 1:30 p.m. – 4:30 p.m. Eduardo Olvera, Senior User Interface Designer, STKU-6 Nuance Communications, Inc. Tuning Speech Applications for Jump-start your knowledge in the field of voice user interface design! This session is designed to quickly get those new to VUI design up-to-speed so Maximum Performance they can make the most of the Principles of VUI Design track at the con- Joe Alwan ference. This tutorial will illustrate why VUI design is the make-or-break fac- tor for speech applications and how to make smart design decisions from 1:30 p.m. – 4:30 p.m. Day 1. Learn how to encourage customers to accept and use speech automation by focusing on the perceptions and reactions of end users STKU-7 throughout the design process. This tutorial will cover the basics in VUI design: writing effective prompts; creating efficient, no-nonsense call flows; VUI, GUI, and MUI: Design Challenges and using evaluation techniques to learn what real customers think. in Multimodal Interaction Catherine Zhu & Julie Underdahl Introduction to Speech Technologies STKU-2 1:30 p.m. – 4:30 p.m. STKU-8 James A. Larson, Vice President, Larson Technical Services User-Centered Evaluation Techniques Designed for attendees new to the speech technology area, this tutorial for Speech Applications provides an overview of today’s key speech technologies. What are the Susan Hura major differences between dictation and conversational recognition engines? Should you use statistical language models (SLMs) or speech grammars, and do they need semantic tags? Should you use concatena- STKU-9 tive or parameter-based engines? Do speaker identifica- Stop Rude Calls: A CCXML Tutorial for tion and verification really work? Is there a need for touchtone recognition VoiceXML Veterans in interactive voice response systems? Who drives the speech dialogue? Moshe Yudkowsky The user, the computer, or both? Where can natural language processing technologies such as natural language recognition, machine translation, STKU-10 response generation, and summarization be used today? Can speech tech- nologies be embedded in mobile devices, or do they require a backend Speaker Identification and Verification server? Should you use POTS, VoIP, or some other communication strat- Judith Markowitz egy? Untangle the voice standards alphabet: VoiceXML, SSML, SRGS, CCXML, PLS, and SCXML.

8 AUGUST 18–20, 2008 w NEW YORK MARRIOTT MARQUIS, NEW YORK CITY SpeechTEK University

THURSDAY, AUGUST 21 VUI, GUI, and MUI: Design Challenges in Multimodal Interaction STKU-7 1:30 p.m. – 4:30 p.m.

Introduction to Grammar Writing STKU-3 Catherine Zhu, Senior Voice User Interface Design Engineer, Intervoice S Julie Underdahl, Senior Voice User Interface Design Engineer, Intervoice p 9:00 a.m. – 12:00 p.m. e

This course focuses on the aspects of design that need to be considered e

Jenni McKienzie, VUI Designer, Travelocity c

Judi Halperin, Speech Engineer, Avaya when creating a multimodal application, such as user modality preference, h This course will arm you with guiding principles and best practices for writ- conflicting mental models, variable usage environments, and technology T limitations. This session will cover the successes and failures discovered E ing GRXML grammars. We will begin with a quick overview of grxml struc- K ture so we can analyze sample grammars. Plus, we’ll examine some cases during the design process for a multimodal application that allowed mobile U when applying best practices really isn’t best (there’s always a loophole!). phone users to order accessories over the phone. The session will also explore the challenges even an experienced GUI or VUI designer faces when n We will consider grammars throughout the design, development, testing, i attempting to design an effective, user-friendly multimodal application. v and tuning of speech applications. Once you have basic grammars down, e r

how can you add cool, advanced stuff to further constrain your grammar s i and optimize recognition? This course is for VUI designers, developers, man- User-Centered Evaluation Techniques t agers, analysts, QA specialists, and others who want to better understand for Speech Applications STKU-8 y how to use grammars to boost performance of their speech applications. 1:30 p.m. – 4:30 p.m. Susan Hura, Principal, SpeechUsability Guerilla Tactics in VUI Design STKU-4 VUI designers and their clients recognize the importance of collecting user 9:00 a.m. – 12:00 p.m. input and including it in the process of designing voice user interfaces to Melanie Polkosky, Human Factors Psychologist/Senior Consultant, IBM speech applications. For many organizations, usability testing has become VUI designers often have to justify their design decisions to clients, peers, a required part of the design process, but usability tests are not always prac- management, and others. But what makes a compelling argument for a tical or even warranted. How do you evaluate a speech application from a user experience design? How do you explain user behavior and prefer- user-centered perspective when usability testing isn’t feasible? This course ences? How do you know what to prioritize? This course will show you how presents a range of methods and techniques for evaluating speech appli- to make the best use of available research or do some research of your cations beyond standard usability testing. Learn how to quickly measure own quickly and under the radar! This session for advanced VUI design- the important aspects of user behavior and perception to make fast and ers will focus on fast but effective techniques for gathering data in social effective improvements to your voice user interface. Discover which evalu- psychology, interpersonal communication, and service delivery as a basis ation techniques will provide the best data for your situation and how to for making design decisions, responding to feedback, and communicat- build continuous opportunities for evaluation into your design and devel- ing with clients. opment process. This course is intended for VUI designers, developers, and managers who want to learn how to make data-based decisions about Natural Language STKU-5 their speech applications. 9:00 a.m. – 12:00 p.m. Stop Rude Calls: Deborah Dahl, Principal, Conversational Technologies STKU-9 Natural language processing is being used in more applications to support A CCXML Tutorial for VoiceXML Veterans increasingly sophisticated speech dialogues. This tutorial will introduce nat- 1:30 p.m. – 4:30 p.m. ural language processing and its role in speech applications. Attendees Moshe Yudkowsky, President, Disaggregate will learn what natural language is; techniques for processing natural lan- Just in time for the elections, here is an application that blocks automated guage; how to use statistical language models (SLMs) and speech gram- telephone calls. This workshop is for people who understand VoiceXML mars with semantic tags for processing natural language; and about emerg- and want to learn about its companion language, CCXML, also known as ing standards and research and their roles in future applications. Attendees voice browser call control. CCXML provides greatly enhanced telephony will have the opportunity to develop tagged grammars illustrating the tech- capabilities and integrates with VoiceXML to provide more powerful appli- niques discussed in the tutorial. cations. In this workshop, we’ll discuss the basic concepts of CCXML and then build a “stop rude calls” application that intercepts incoming phone Tuning Speech Applications calls to determine if they’re from a live person. Bring your laptop and take for Maximum Performance STKU-6 home the “stop rude calls” application for possible extension into a call routing or answering machine. 9:00 a.m. – 12:00 p.m. Joe Alwan, VP/GM AVOKE Caller Experience Analytics, BBN Technologies Speaker Identification and Verification STKU-10 Methods for tuning speech applications have grown and evolved within the 1:30 p.m. – 4:30 p.m. industry. Just as holistic views of the application replaced narrow efforts focused only on the recognition engine, the newest approaches step back Judith Markowitz, President, J. Markowitz, Consultants even further from the engine and examine caller success from dialing to Join us for an informative session covering the A–Z of speaker biometrics. hangup. This course will teach you how to manage a successful tuning We’ll look at technology, applications, and markets. Along the way we’ll expe- project to get maximum value from your speech applications. Topics include rience what it is like to use the technology and answer questions such as setting tuning objectives, understanding caller behavior and intent, meas- these: How do speaker biometrics work? Can a tape recorder fool a speaker- uring disatisfiers and inefficiencies, and identifying the highest impact tun- verification system? Can a professional mimic fool these systems? How do ing opportunities. Using actual case studies and class exercises, the course speaker biometrics compare with other biometrics? Who is using it and why? will cover a framework and methodology that you can use to manage both How should I explain the system to my customers/employees? This session in-house and vendor-partner tuning projects. presents no-nonsense concepts and real-life examples of what works!

BREAK 12:00 p.m. – 1:30 p.m.

WWW.SPEECHTEK.COM 9 Planning for Speech A102 Keynote 11:15 a.m. – 12:00 p.m. MODERATOR: Nava Shaked, CRM and CC Practice Leader, IBM

8 Your organization is making the leap into speech technology—congratulations!

1 Speech in the Mainstream But what happens next? This session will arm you with information about steps

T and Beyond you can take to prepare for a speech project’s development. Learn what data S 9:00 a.m. – 10:00 a.m. Broadway Ballroom, 6th floor will ease the transition to speech and what team resources will be needed to

U w Ray Kurzweil, author of The Age of Spiritual Machines: When keep a speech project on track. Learn how defining success criteria up front G Computers Exceed Human Intelligence and The Age of Intelligent can provide a framework for decisions throughout the project and drive a pos- U Machines itive experience from the start. Speakers in this session will help you ensure A that you’re ready for speech when your project begins. , We’re living in a world of rapid technological innovation and increas- Y ing pervasiveness of information technology. Consumers not only A Speech Is Coming: What Do I Do Now? appreciate these advances in their professional and personal lives, D Catherine Zhu, Senior Voice User Interface Design Engineer, Intervoice they are demanding it. What does this mean for those who are driv- N The decision to move to speech has been made, and now it’s just a mat- ing development and acceptance of speech technology? Glean

O ter of logistics before the project can officially start. How can you profitably insight from one of the most distinguished speech technology inno- use this time? Learning speech technology basics, beginning interdepart- M vators of our time, Ray Kurzweil, who will share his views on how mental communication, and gathering data about your business and poten- speech technology will work in conjunction with other emerging tech- tial callers are just a few factors to consider for managing a smooth transi- nologies to bring us to an age of intelligent machines. tion. This session will explore proactive steps to take as a company when you know speech is coming. Preparation Leads to Success TRACK A Soho Complex (7th Floor) w Caroline Nelson, Speech Solutions Team Technical Lead, Nortel MEETING BUSINESS GOALS What do customers need to know in preparation for a speech application WITH SPEECH TECHNOLOGY deployment? An awareness of the caller base and a comprehensive under- Mainstream Speech 2008: standing of your back-end data are just the beginning. Other elements, A101 such as involving the right players and setting success criteria at the start The State of the Industry of the project, also play a huge role. A detailed discussion of these items 10:15 a.m. – 11:00 a.m. and more add up to the top 10 tips for deploying speech. MODERATOR: Tim Moynihan, Vice President, Global Marketing & Sales Support, Envox Speech is a mainstream technology, used by millions across the world every SPONSORED KEYNOTE LUNCH w Sponsored by day to get information, perform transactions, and manage their daily lives. Phone 2.0: The Mobile Generation at Work What do end users really think of speech technology? What are the hot but- 12:00 p.m. – 1:15 p.m. Broadway Ballroom, 6th floor tons in the speech vendor community? What do organizations using speech w Beatriz V. Infante, President and CEO, VoiceObjects, Inc. have to say about the technology? Panelists in this session have completed Phone self-service has evolved with today's new generation of ambitious client, vendor, and end-user surveys of the speech industry. The mobile users who expect a much higher level of phone self- resulting data can tell us a great deal about the current state of the speech service than the common one-size fits all menu structures. The industry and where we need to be tomorrow to continue to grow and thrive. Come with questions and be prepared for a spirited debate of the findings presentation shows how Phone 2.0 is now becoming the new on the state of speech in 2008. standard for phone self-service with companies today and how consumers are now enjoying personalized self-service experiences over the phone, Year 2 Speech in the Mainstream: just like they have over the Web for many years. In addition, new phone Stakeholder Views From 360 Degrees channels are being adopted by many companies for text, multimedia and Tim Pearce, Global Solutions Manager, Self Service, Dimension Data mobile Web interactions. Mike Bergelson, Director, Product Management, Cisco Are vendors and end users on the same page about speech recognition Preventive Medicine: applications? This discussion picks up from where SpeechTEK 2007 left Keeping Your Speech System Healthy A103 off, showcasing the results of the global Alignment Index, which tracks the 1:15 p.m. – 2:00 p.m. views of 1,200 speech-app consumers and 128 speech vendors to see how MODERATOR: Sunil Issar, Director, Convergys closely they’re aligned. Your speech application is up and running—now you can relax, right? VoiceXML and IVR Adoption for Self Service Portals: Wrong! Speech applications are dynamic entities that require lots of ongo- The European Mobile Operators Perspective ing TLC to perform at peak effectiveness. Experts in this session will teach Monique Bozeman, Independent Marketing Consultant, Monique Bozeman you how to monitor and manage your speech application throughout its Consulting lifecycle and how to maintain excellent performance without breaking the Bonnie Crater, Senior VP Marketing, VoiceObjects budget. If you currently have a speech application, this session will tell you According to research by Datamonitor, VoiceXML revenues and port ship- what you need to know to keep it on track for the long haul. ments in both North America and EMEA surpassed those of traditional IVR Now What? Managing and Maintaining Speech Deployments for the first time in 2007. VoiceObjects recently sponsored research to “take Fran McTernan, Speech Operations Lead, Nortel the temperature” of the European Union’s highly competitive market of mobile network operators. This presentation will highlight those survey find- Businesses commonly turn to technology providers and vendors to get their ings, including operators’ customer service challenges, their plans for initial speech deployments up and running. But once the system is earn- improving customer satisfaction, and how they will differentiate themselves ing its keep, how should you maintain it? What skills does a team need to through self-service offerings as they move from the customer acquisition keep up with changing and growing business needs? This session is to customer retention phase in a rapidly consolidating market. geared toward IT managers responsible for managing and maintaining speech systems, focusing on the high-level skills and knowledge required to do so.

10 AUGUST 18–20, 2008 w NEW YORK MARRIOTT MARQUIS, NEW YORK CITY MONDAY,MONDAY, AUGUSTAUGUST 1820

Strategies to Cost-Effectively Maintain Understanding the Impact of Customer Self-Service Phone Portals Experience on Your Business Ingo Bors, Industry Consultant, Ibostar Consulting Services CCS Glen Graham, SVP, Business Operations, Bank of America Self-service phone applications often go unmodified due to increasingly This presentation will explain how Bank of America examines customer complex IT and business process requirements. The result is a limited abil- experiences across multiple channels (IVR, agent, web, etc.) to identify the M ity to adapt voice self-service dialogues in real time based on a caller’s exact events resulting in customer confusion and frustration. Learn how O behavior and a restricted flexibility to accommodate change requirements. Bank of America uses this information to prioritize enhancements that max- N

This presentation will discuss business conditions driving modifications to imize user adoption and customer retention. D

self-service phone portals, presenting examples of how to use a modular A approach and services-oriented architecture (SOA) for maintaining them. A Voice Solution for ATM Using TTS Y Ricardo de Carvalho Destro, Technology Director, Volans Technology , CCS A

Agile Speech Projects A104 A TTS based solution for ATM equipments will be presented in this session. U 2:15 p.m. – 3:00 p.m. This solution aims to allow blind customers (or with low vision), to use ATM G

machines for banking transactions as withdraws, balance etc. This multi-plat- U MODERATOR: Phillip Hunter, Vice President, Interaction Design,

form solution was designed to support many languages and to be customized S SpeechCycle with low impact of development in the Bank’s transactional applications. T Agile software development is gaining wider acceptance for the benefits it 1

offers over traditional waterfall development methods. Agile development 8 A106 allows an organization to have code up and running more quickly and to The Business of Speech Analytics be more responsive to changes by developing software in discrete chunks 4:15 p.m. – 5:00 p.m. that can be easily modified. How do agile development processes impact MODERATOR: Dan Miller, Senior Analyst, Opus Research speech projects? Is it possible to design a VUI that fits within the constraints of agile development? Speakers in this session explore the benefits and Roundtable: Building Better Applications challenges of agile development for speech projects. and Driving Business Value With Analytics PANELISTS: Adopting Software Engineering Practices for VUI Development Michael Codini, CTO and Co-Founder, VoiceObjects Charles Lewis, Senior Technical Trainer, Bloomberg, LP Greg Borton, VP of Analytics, Nuance Communications As call flows become more complex, the discipline of software engineer- Cliff LaCoursiere, Senior Vice President of Business Development, CallMiner ing can offer guidance for the creation and maintenance of robust, scala- As self-service phone applications mature and become broadly deployed ble products. This session will describe the successful application of stan- to reduce costs and increase customer self-service, there is a critical need dard software development practices, including source-code control, for companies to understand how their customers are using these systems. coding standards, code walk-throughs, and the application of object-ori- Despite the tremendous interest in business intelligence and analytics ented programming principles to the development and ongoing improve- among mainstream IT, the call center and speech industries remain woe- ment of highly complex call flows. The application of these processes and fully behind. This panel will investigate the role of analytics in three key areas practices to VUI development presents unique challenges that will be dis- of call center technology: self-service, speech, and customer care. Atten- cussed. dees will learn how new business intelligence and analytics technologies are helping customers build better applications, improve customer service, Flexible Design in Support of More Rapid Deployment and drive business value. Leslie Carroll Walker, Staff Engineer, VUI Design, Intervoice As customers move away from waterfall methods of deployment, design- WELCOME RECEPTION w Sponsored by Angel.com ers may need to have several different design initiatives under way simul- 5:00 p.m. – 7:00 p.m. in the Exhibit Hall taneously. This presentation will look at some practical approaches to the design process in support of more rapid deployment models. Documen- tation options, project management options, and other workarounds will TRACK B w Empire Complex (7th Floor) be offered and discussed. This presentation will be particularly useful for VUI DESIGN PRINCIPLES AND TECHNIQUES consulting designers who are working on large, ongoing initiatives, but should also prove helpful for smaller initiatives and in-house designers. What End Users Really Want B101 10:15 a.m. – 11:00 a.m. A105 Speech in Financial Services MODERATOR: Juan Gilbert, T-SYS Distinguished Associate Professor, 3:15 p.m. – 4:00 p.m. Auburn University MODERATOR: Ken Rehor, Founder, VoiceXML Forum In speech projects, requirements are sometimes thought of as a laundry Banks and financial institutions are adopting speech technology. This ses- list of items to be handed from the client to the vendor building the speech sion presents two different uses of speech technology. Bank of America application. Because clients don’t know VUI, requirements are sometimes describes how it uses analytics to identify customer behavior and experi- less than sensible. Moreover, requirements rarely include the perspective ence to improve customer treatment and ultimately increase financial ben- of the application’s end user. Therefore, building an application on such efit. And Volans Technology describes how it uses speech synthesis tech- requirements often produces applications that are poorly adopted by users nology to assist low-vision customers in using ATMs in any of several and thus fail to meet client goals. Experts in this session will explore why languages. requirements assembled in a vacuum are insufficient and discuss meth- ods for discovering truly meaningful requirements. The Big Gap: What Customers Want Breakfast & Coffee Breaks Versus What Customers Get Moshe Yudkowski, President, Disaggregate A continental breakfast will be provided for conference attendees When someone calls your office and you’re not there, he might want to know when you’ll be back, whether you received a file he sent you, and each morning from 8:00 a.m. to 9:00 a.m. before the keynote session. what you thought about his proposal. What this person gets, however, is Additional coffee breaks will take place each morning and afternoon. something entirely different: voicemail. In this session, we’ll discuss the Please check the schedule for exact times and locations. huge gap between what customers want and what customers get, and offer ideas on how to create services that actually deliver what they want.

WWW.SPEECHTEK.COM 11 Unearthing the Real Requirements You also will be encouraged to challenge yourself and your vendors to make Jenni McKienzie, Senior Business Solutions Advisor, Travelocity courageous design decisions that will empower customers with excellent VUI designers often receive puzzling requests from clients. Valid reasons self-service. typically drive their requests, but the businesses haven’t come up with the It’s OK to Tell Clients They’re Wrong 8 right solutions. We’ll dissect several real requests for IVR modifications to

1 Jessica Stevens, Senior Engineer, VUI Design, Intervoice uncover their underlying problems, and then we’ll examine better alterna- Jenny Burr, Manager, Speech Technology, Global Consulting Services, T tives than the ones proposed. Finally, we will look at how to go back to these

S Intervoice clients and sell them on your plans.

U Dealing with customers is a large part of our jobs in VUI design and tun-

G ing. We have all run into differences with our customers, and both our suc- B102

U Managing Client/Vendor Relationships cess stories and cautionary tales can be informative. This session will pro-

A 11:15 a.m. – 12:00 p.m. vide real-world examples of trying to find that sweet spot between business ,

Y MODERATOR: Darla Tucker, Principal, Global Consulting Services, and caller needs, demonstrating the importance of making the effort to rep-

A Intervoice resent our expertise in the industry and reminding clients why they hired

D VUI designers are the keepers of a great deal of highly specialized knowl- us in the first place.

N edge about speech technology and the way users interact with it. While

O VUI designers are hired specifically for this specialized knowledge, clients Sponsored by regularly challenge VUI specialist design decisions and insist on prompts SPONSORED KEYNOTE LUNCH w M and call flows that make VUI designers cringe. This session is about find- Phone 2.0: The Mobile Generation at Work ing the balance between providing expert guidance to clients and accept- 12:00 p.m. – 1:15 p.m. w Broadway Ballroom, 6th floor ing requests that will compromise the design. Learn how to decode client Beatriz V. Infante, President and CEO, VoiceObjects, Inc. requests to understand the real intent, include clients in the design process, and gently steer them toward solutions that will meet their ultimate goals. Reverse Navigation Hands-On Session B103 Caution Ahead: Voice User Interface Is a Two-Way Street 1:15 p.m. – 2:00 p.m. David Martin, Principal Consultant, Speech Engineering, Avaya MODERATOR: Jenni McKienzie, Senior Business Solutions Advisor, The goal of this presentation is to facilitate critical thinking about deploy- Travelocity ing user-centric business solutions. The discussion will cite recent applica- Peter Krogh, Speech Mechanic, SpeechCycle tions in which business processes had the potential to detract from the user How do you navigate back to a previous state in a VUI? Unlike webpages, experience, focusing on how a well-designed user interface accounts for VUIs have no “back” button that transparently takes users back a single both business and customer goals and how the attitudes of business stake- step in the interaction. Does a simple speech analog to the back button holders and VUI designers can affect the success of a speech application. make sense in a VUI? How, and when, should this functionality be offered

12 AUGUST 18–20, 2008 w NEW YORK MARRIOTT MARQUIS, NEW YORK CITY MONDAY, AUGUST 18 to users? Should reverse navigation be universal or are specific instantia- leaves users perplexed, annoyed, and wondering why they can’t get there tions required for different applications? Explore the nuances of reverse from here. Learn why interface symmetry is desirable, as well as how to navigation in this hands-on session. After a brief introduction, be prepared plan for it, champion the cause, and educate both callers and clients as to to work together in small groups to explore this issue and to present solu- its possibilities and limitations. tions to the audience. M

How May I Help You? B106 O B104 Universal Commands 4:15 p.m. – 5:00 p.m. N

2:15 p.m. – 3:00 p.m. MODERATOR: Elizabeth Strand, Director of User Experience Design, D

MODERATOR: Charles Galles, Speech Scientist, Multimedia Applications, Tellme A Nortel Networks Y

Statistical language models allow us to pose open-ended questions and ,

Is there a set of commands that all speech applications should share? If allow users to give more free-form responses. This sounds simple, but for A we offer users help throughout a speech application, does it do any good? “How may I help you?” questions, the devil is in the details. The exact for- U This session explores the highly debated issues of whether we need uni- mulation of the question can have a huge impact on users’ responses, and G versal commands across applications, determining what these commands conversely, knowing how users respond can help craft more effective open- U should be, and whether universal commands truly benefit users of speech S

ended questions. In this session, we explore the details of writing open- T applications. Join us as two VUI practitioners share evidence for how to ended prompts and analyzing user responses to these questions. handle universals in VUI design. 1 People Say the Darnedest Things 8 Should Universal Commands Be Universal? Charles Galles, Speech Scientist, Multimedia Applications, Nortel Rita Dhruve, Speech Solutions Specialist, Nuance Networks A small set of commands is becoming a de facto standard for universal Open prompts (such as “How may I help you?”) produce interesting commands in VUI design. However, guidelines for how and when to offer responses. This discussion analyzes responses based on caller profiles universal commands, and what response should be expected of the sys- and VUI design issues across several applications. The data presented will tem, don’t exist. This session will present examples from current speech help you understand how to deliver an excellent caller experience, drive applications to demonstrate how to successfully implement a set of univer- customer loyalty, and strengthen your brand. sal VUI commands. Natural Language Questions for Technical Support Does Universal Help Help? Applications: A Winning Strategy James Mesbur, Voice Interaction Designer/Speech Scientist Engineer, Mary Constance Parks, Senior VUI Designer, Nuance Communications SpeechCycle Karen Moylé, Voice Interaction Designer, SpeechCycle For a recent technical support application, the goal was to improve the Universal commands are intended to give a caller control over a speech accuracy of an existing natural language question and back-off menu. Three interaction and provide a consistent approach to tasks. But universal help, different designs were created to challenge the existing flow, one of which as it is typically implemented, has never delivered on this promise. As a significantly improved both accuracy and automation. The characteristics result, we have implemented a set of context-sensitive help options, offered of the winning design will be presented, along with takeaways for design- only when extra help makes sense. We’ll discuss how this approach allows ing technical support applications and natural language questions. for simpler grammars, lets callers educate themselves when they feel it’s necessary, and results in a marked increase in dialogue object success, WELCOME RECEPTION w Sponsored by Angel.com call completion, and caller satisfaction. 5:00 p.m. – 7:00 p.m. in the Exhibit Hall

Lessons in Multimodal Usability B105 3:15 p.m. – 4:00 p.m. TRACK C w Majestic Complex (6th Floor) MODERATOR: Catherine Zhu, Senior Voice User Interface Design ADVANCED SPEECH Engineer, Intervoice TECHNOLOGIES SYMPOSIUM Applications using speech and visual modalities have arrived and present special challenges to user interface designers. Speech and visual modali- Advanced ASR: A Global Perspective C101 ties ideally work in concert in multimodal applications, allowing the user 10:15 a.m. – 11:00 a.m. flexibility in information input and output. When multimodal goes wrong, however, speech and visual modalities can interfere with each other, com- MODERATOR: Tom Schalk, Vice President, Voice Technology, ATX peting for the user’s attention and making multimodal less effective than Roberto Pieraccini, Chief Technology Officer, SpeechCycle unimodal applications. Speakers in this session share their experiences in This session provides a global overview on the state of research in the many multimodal design and give recommendations for helping multimodal appli- areas related to speech and language, such as speech recognition, lan- cations live up to expectations. guage understanding, spoken dialogue systems, speech summarization, speech-to-speech translation, audio indexing, speech synthesis, speaker Multimodal Speech Usability Lessons verification, and supporting technologies. Learn about the wider speech Eduardo Olvera, Senior User Interface Designer, Nuance Communications technology research community, the conferences, the organizations, and Multimodal user interfaces can combine visual-, touch-, gesture-, and loca- the publications. Finally, learn about breakthroughs in advanced technolo- tion-based features. But what happens when you add speech recognition gies where academic research is currently making progress and their poten- capabilities to designs that are so new they lack a clear set of usability guide- tial impact in the commercial world. lines? Based on actual multimodal speech usability testing, this session will discuss the challenges of multimodal design and provide recommenda- Application of Advanced ASR C102 tions for successful implementations. 11:15 a.m. – 12:00 p.m. You Can’t Get There From Here, But Why Not? MODERATOR: Tom Schalk, Vice President, Voice Technology, ATX Matt Prather, Staff Engineer, VUI Design, Intervoice This session focuses on speech recognition technology used to complete In an increasingly multimodal world, users expect a voice interface to do challenging tasks such as automated directory assistance, address entry, the same things, in the same ways, with the same results, as the desktop understanding newscasts, and the handling of conversational speech in and web-based interfaces they’re already using. But, all too often, function- mobile devices. In addition to exploring key advances in core speech recog- ality can be radically different across a company’s various interfaces, which nition technology, performance results from real-world deployments will be WWW.SPEECHTEK.COM 13 14 MONDAY, AUGUST 18 uted mobileenvironments. and ne novel andchallenging usecasessuch asweb search andvoicemail-to-text; t deplo ing the unprecedented quantitybeinggenerated ofuserdata by current explore factors shapingthe next generation ofspeech technology, includ- use ofs overcoming becomesever more language barriers prominent. Increased As t MODERATOR: 1:1 Success WithMachineT Beatr 12:00 p.m.–1:15 p.m. Phone 2.0:TheMobileGenerationatWork SPONSORED poised t next few promise years to beeven more dramatic,with speech interfaces resulted inanimpressive expansion ofspeech applicationcapabilities.The Steady advances incore recognition technology over the decadehave past Vlad Sejnoha, Performance ofAdvancedSpeechTechnology andvalidation. analysis document classification,andsearch query detection, phoneticindexing andsearch, language/semantic technologies, syndication. Technologies discussedincludeaudio/speech, voice andmusic c length broadcast (longform) to single segmentscontaining stories (web and semantictechnologies to from anewscast automatically itsfull- convert concrete, deployed successfully applicationthat appliesspeech, language, syndication, anddiscoverability ofonlinevideos.Thesessionwillpresent a c manage andnavigate such audiovisual content. Thispresentation willdis- The current explosive growth ofonlinevideonecessitates improved ways to Gavalda, Marsal Streamline VideoSearchandEditing Using PhoneticAudioSearchTechnology to terns, andwhere the technological gapsarise. presented to andanalyzed show trends inautomation rates, usage pat- from to couplingspeech recognition andunderstanding create solutions AUGUST 18–20, 2008 he drive toward lowering applicationdevelopment andmaintenance cost; uss the role that semanticcomputing canplay inthe editing,publishing, lips) andprepares such clipsfor aware onlinepublishingandsemantically w w w w t session breaks, lunches, receptions, andinformal gatherings ofcus- to meetother customers, includingbreakfasts,ing opportunities In additionto formal conference sessions,there are many network- f Listenvariety ofindustries. to presentations from companies inthe andspeakersolutions, analytics, authentication across awide successfull s applications, andhow they avoid commonpitfalls. Customer case )give you unparalleled information onhow speech t icon( Study Customer presentations casestudy marked with aCustomer Case Customer CaseStudies By CustomersFor ollowing industries: ollowing industries: omer ec 5 he importance of timely accessto oftimely information grows,he importance the necessityfor tudies willbepr Financial Services Government/Utilities Healt CPG/R iz V p.m. –2:00 yments; t hnologies ar w t o atistically basedcomputationalmethodsatistically andtechniques evolved s . demands onASRfle jum Inf wit hcare et ant h p y ail Vice President,ChiefScientist, Nuance Communications Bill Scholz, he dif similar int C t pr e, o C Director ofIncubation,Nexidia ocur ne S e KEYNOTE President andCEO,V esent r w f eall er e, implement, anddeploy existing speech levels ofpower andusability. Thispresentation will ence betw y erests. ed onaut being applied.Presenters share how they w President, NewSpeech LLC Broadway Ballroom, 6th floor xibility stemming from deployments distrib- ranslation LUNCH een core anduser-perceived accuracy; omatic speec w w w w oiceObjects, Inc. Order Entry A Insurance Telecommunications w utomotive w Sponsor E OKMRIT MARQUIS MARRIOTT YORK NEW h r ecognition ed b y C103 improve machine translationandinformation extraction components. cross-lingual information propagation to andnamespellingnormalization technical challenges inbuildingthe TALES platform with anemphasis on to English, andinformation extraction. Thispresentation shows someofthe ture, automatic speech-to-text conversion, machine translationofforeign text guage ExplorationSystem, codenamedTALES. TALES videocap- performs t IBMmadeabigstepeign mediainreal timewithout language barriers. People are interested searching ineasily andmonitoring awiderange offor- Imed Zitouni, Translingual AutomaticLanguageExploration applications willalsobediscussed. tunities for these andintelligence pairing technologies incustomer support New oppor- success to ensure usabletranslationsofspoken information. monitoring andspeech-to-speech translation,alongwith requirements for ing speech andautomated translationsolutionsfor videobroadcast munication across multiplelanguages. Thispresentation willlookatexist- els inrecent anditisnow years, possibleto deliver real-time humancom- Speech technology andtranslationtechnology have advanced to new lev- Hannah Grap, A Speech Technology andAutomatedTranslation: r Through ofconference aseries callsandworkshops,services. we have been research and new methods have beendeployed inimpressive and trials in menu-basedorfinite state machines.) Thishasbeenan area ofactive as asystem endowed with ahigherlevel ofreasoning than exists typically dards to (Wedevelopment define“advanced” ofadvanced speechservices. The VoiceXML Forum Tools Committee isexploring ways to introduce stan- David Thomson, Architectures forAdvancedDialogues some ofwhic flexible interactions to allowing callers request hundreds ofunique tasks, grammar, data, robust and powerful andcodeinfrastructuresthat support ing andmultipleo adapt accor sim and callerdemandsmeanenablingbr broad (callrouting andstocks)ing) orvery contexts. Butwhencustomer applications characterized deep(insuranceortroubleshoot- by either very fewThe last have years seengrowth inthe needfor complex highly speech Phillip Hunter, Managing DynamicContextual Complexity More ThanCallSteering: r Speec MODERA 2:1 Advanced Dialogue’sGrowingSophistication in sophis will outlinepr pr reduce the development The for cost buildingadvanced speech services. dar by amajor standards organization. supported dialogue services design andlooks atthe evolution toward anintegrated modelfor advanced This sessione lizing inference andreasoning capabilityborrowed from the AIcommunity. robust andflexible interaction—evencode infrastructuresthat support uti- T ments for troubleshooting, callrouting, information anddetailed retrieval. successes inmachine translation. explores keyresearch inthisarea andprovides cogent examples ofreal speech-to-speechable to perform translationinnearreal time.Thissession oward this goalwith the development ofthe Trans-lingual Automatic Lan- eq e o vie esentation willreviewesentation advanced dialogues and methods for supporting Winning Combination meet t uir ds canbeused.Bystandardizing system components, we hope to help plicity ofaccess,int 5 wing ar p.m. –3:00 ing bo h applications continueto grow incomplexity andsophistication, tication, r T hese needs,t OR: ding t c h hit oposed architectures andstandards underconsideration. h , Sr. Research Scientist, IBMResearch the deepeningandbroadening ofcontext to meetthe require- xplor Marketing CommunicationsManager, Language Weaver Bill Sc Vice President,Interaction Design,SpeechCycle in E OKCITY YORK NEW ectur l y. Highlighted in acustomer deployment involving callsteer- volve and contexts. dozensofturns variously arranged CT eq t her applications,SpeechCycle discussesadvances in es ins O, SpeechPhone e uir holz, erface and application design and construction must andapplicationdesignconstruction erface options andar he dialoguedesignpr ing t t ances oft President, N he integration of powerful data, grammar, data, he integration ofpowerful and his g e oadly moreoadly complexity with increased e e xplor wSpeec rowing indialogue sophistication ocess hasg ing whatnew orexisting stan- h LL C rown significantly C104 MONDAY, AUGUST 18

cussed in the presentation by illustrating real customer problems and issues C105 Natural Language Processing Techniques with some solutions suggested. In addition, the presentation will address reg- 3:15 p.m. – 4:00 p.m. ulation and business considerations that result from this approach. MODERATOR: Tom Schalk, Vice President, Voice Technology, ATX

This session will examine robust linguistic dialogue tools applicable to voice WELCOME RECEPTION w Sponsored by Angel.com M application development. Proper use of these newly developed tools makes 5:00 p.m. – 7:00 p.m. in the Exhibit Hall O

it easy to realize a cost-effective implementation of caller query handling N

and executing transactions in a natural language fashion. Additionally, this D session will discuss adaptive techniques designed to dynamically adjust Shubert Complex (6th Floor) TRACK D w A prompting in terms, speaking rate, and content. Ultimately, applying the DEVELOPMENT AND DEPLOYMENT Y methodology and products discussed in this session results in an improved , D101 A user experience, better automation rates, and increased IT efficiency. Design Methodologies and Tools U

10:15 a.m. – 11:00 a.m. G Improved Customer Experience with a Natural Language Solution MODERATOR: Ingmar Kliche, Project Manager, T-Systems Enterprise U Services GmbH S

Peter Trompetter, VP Global Development, GyrusLogic, Inc. T The presentation will emphasize the benefits of ASR/VoiceXML applica- How can systematic design methodologies and usable tools decrease the time and effort to create useful user interfaces? The first presentation 1 tion developments with the use of robust artificial intelligence (AI) tools for 8 easy and cost-effective implementation of answering a caller’s questions, describes developer experiences designing a multimodal user interface executing transactions in a conversational dialogue and/or natural language using a standard product development methodology, modified to include fashion. The solution presented is a linguistic AI conversational dialogue special steps for designing voice and graphical components. A formal pro- product suite to complement the ASR and/or VoiceXML application devel- cedure for evaluating and selecting voice-enable features to include with opments achieving the caller’s first contact resolution. the GUI will be detailed. The second presentation overviews speech appli- cation development tools, identifies gaps created by using disjointed tech- Adaptive Audio IVR Software for IT Efficiency nologies, and describes the need for a new generation of tools to reduce Daniel O’Sullivan, President/CEO, Interactive Digital the overall complexity of delivering speech applications and to strike the Adaptive audio IVR software can dynamically adjust the speaking rate (in right balance between customer controls over an application while main- words per minute) and audio message content of voice applications on taining integrity. individual caller skills. This personalizes the call experience as it happens, A Multimodal Interface for Call Center Agents creating more responsive and more productive customer experiences. The Matthew Yuschik, Human Factors Specialist, Multichannel Self Care, process emulates what humans do naturally to communicate more effec- Convergys tively with each other during normal conversation. The benefits of adaptive audio include decreased average handle times and increased average han- How can systematic design methodologies decrease the time and effort to dle rates, ultimately resulting in increased customer satisfaction. create useful multimodal user interfaces? This presentation describes devel- oper experiences designing a multimodal user interface using a standard C106 design methodology, modified to include special steps for designing the Virtualization and Intelligent Integration voice and graphical components. This presentation describes a formal pro- 4:15 p.m. – 5:00 p.m. cedure for evaluating and selecting voice-enable features to include with MODERATOR: Tom Schalk, Vice President, Voice Technology, ATX the GUI and discusses product development stages and test results. The first presentation identifies how virtual IVR technology can enable each deployed speech application to operate in an independent space It’s All Connected: The Dysfunctional Relationship Between with its own speech resources, configuration, and reporting. Virtualization the Toolbox and Successful Speech Applications is ideal for large companies, carriers, hosting, and any environment with Leah Eyler, Product Manager, SpeechCycle many applications. This solution simplifies the deployment of multiple Today speech application developers have a mixed bag of tools targeted speech recognizers, development tools, applications, and teams of devel- at a range of skills, resulting in a range of solutions but very few in support opers. The second presentation describes how speech technology of a comprehensive approach to development. This presentation will dis- increases safety and improves communication for police officers who must cuss the evolution of speech application development tools and the gaps interact with a license plate database while driving or interviewing sus- created by using disjointed technologies. Because of the connections pects outside of the patrol car. between VUI, development, QA, and analysis, there is a need for a new generation of tools to reduce the overall complexity of delivering speech IVR Virtualization: Platforms for Parallel applications and to strike the right balance between customer controls over Speech Development and Deployment an application while maintaining its integrity. Learn what to consider when Rob Kassel, Vice President, Marketing & Product, Holly Connects selecting development tools. Virtual IVR technology enables each speech application deployed to a plat- form to operate in an independent space with its own speech resources, Speaking and Listening to Mobile Devices D102 configuration, and reporting. Virtualization is ideal for large companies, car- 11:15 a.m. – 12:00 p.m. riers and hosting, and any environment in which there are many applica- MODERATOR: Ahmed Bouzid, Senior Product Manager, Angel.com tions because it provides a natural efficiency for developers, user interface designers, and operational staff. Virtualization simplifies the deployment of Speech technologies will be an integral part of the user interface on multiple speech recognizers, multiple development tools, multiple applica- cell phones, PDAs, and other mobile devices. This session discusses the tions, and teams of developers. problems of designing and implementing user interfaces for mobile devices which are expected to soon be widely used. The Challenges of Mixing Speech Technologies in the Development and Deployment of Speech Applications Steps to Determine Multimodal Mobile Interactions Matthew Yuschik, Human Factors Specialist, Convergys Nava Shaked, CRM and CC Leader, IBM We are often asked by our customers both to enhance system efficiency and By observing call center agents using a multimodal tool that overlays to maximize results. These goals can be reached by combining two or more speech onto an existing GUI, multimodal procedures are identified that speech technologies such as speech recognition and speech biometrics. reduce the complexity and increase the efficiency of the transaction. Successful cross-technology integration requires not only the core technol- This process becomes an early step in migrating an agent-enabled ogy interface but also development of the appropriate VUI. These will be dis- transaction in the call center to a self-service transaction performed WWW.SPEECHTEK.COM 15 by the caller on their mobile hand-held device. Yuschki describes the steps Modular Data Components in SCXML in the design procedure that highlight how call center agents become a James Barnett, Senior Technical Director, Aspect Software testbed for end-user multimodal UIs. The State Chart XML (SCXML) describes XML syntax for representing dia- logue flow control using state chart notation (an extension of state transi-

8 Sponsored by tion diagrams). SCXML will be an important component of the forthcom- 1 SPONSORED KEYNOTE LUNCH w ing VoiceXML 3.0 as well as be used for a variety of nonspeech applications.

T Phone 2.0: The Mobile Generation at Work The latest W3C working draft of the SCXML specification provides a mod- S 12:00 p.m. – 1:15 p.m. Broadway Ballroom, 6th floor ularization of the language. This talk will provide an overview of this mod- U w Beatriz V. Infante, President and CEO, VoiceObjects, Inc. ularization with particular emphasis on its pluggable data model. Specifi- G cally, platforms can plug in different data languages, such as XML and U

A New Standard Languages for ECMAScript, to suit different applications. , D103

Y Developing Speech Applications D104 A 1:15 p.m. – 2:00 p.m. Advanced Techniques for Using Grammars

D MODERATOR: Yves Normandin, President, NuEcho 2:15 p.m. – 3:00 p.m. N The World Wide Web Consortium (W3C) has published a working draft for MODERATOR: Rob Kassel, Vice President Marketing & Product,

O two new standard XML languages for developing speech applications. The Holly Connects

M Pronunciation Lexicon Specification is a standard XML interface for speech This session discusses two advanced grammar techniques. First, the W3C recognition engines and speech synthesis engines to access lexicons of standard ABNF grammar format is extended for the design and implemen- words and their pronunciations. The State Chart XML (SCXML) describes an tation of dynamic grammars that can be edited, tested, and debugged using XML syntax for representing dialogue flow control using the popular state the same set of tools that help a speech scientist build effective static gram- chart notation. SCXML will be an important component of the forthcoming mars. Second, when standard transcribed data is not available, algorithms VoiceXML 3.0 as well as in a variety of nonspeech applications. generate grammar weights based on data available from the application. These presentations will cover how to use both innovative techniques that How to Improve TTS and ASR Performance promise to improve speech recognition accuracy. Paolo Baggia, Director of International Standards, Loquendo The accurate specification of the correct pronunciation is critical to the suc- Developing Robust and Efficient cess of many speech applications, especially when targeting multilingual Dynamic Speech Recognition Grammars applications. For example, the incorrect pronunciation of proper names and Dominique Boucher, Lead Software Developer, NuEcho place names can confuse and mislead users. The Pronunciation Lexicon The W3C standard ABNF grammar format is extended for the design and Specification (PLS) is a new W3C voice browser standard designed to enable implementation of dynamic grammars. Using this approach, dynamic gram- interoperable specification of pronunciation information for both ASR and mars can be viewed as a seamless evolution of existing static grammars TTS engines. The presentation will demonstrate how PLS enables the effi- instead of a separate set of resources in a project. These dynamic gram- cient development and deployment of speech applications that share pro- mars can be edited, tested, and debugged using the same set of tools that nunciation lexicons. Extensive examples and case studies will be presented. helps a speech scientist build effective static grammars, including a

Don’t let technology come between you and your customers!

PHONE INTERNET EMAIL FAX CHAT

Proactive Testing for Confident Customer Interactions Test and monitor your multi-channel contact center solutions with IQ Services today! IQ Services oers multi-channel, customer experience testing services for unied communications solutions and contact center technologies. Since 1996, IQ Services has delivered value and unparalleled customer service to a diverse clientele ranging from small business owners to Fortune 100 corporations.

IQ Services’ sophisticated tools and methodologies allow us to tailor a variety of services to meet your unique business needs and integrated testing requirements.

To learn more about our managed testing services, please check us out at www.iq-services.com. Also, visit us at Booth #408 and the Multimodal Lab on August 20th.

6601 Lyndale Ave South Suite 330 t Minneapolis, MN 55423 Tel: 612-243-6700 t Fax: 612-869-6200 www.iq-services.com t [email protected]

16 AUGUST 18–20, 2008 w NEW YORK MARRIOTT MARQUIS, NEW YORK CITY MONDAY, AUGUST 20 feature-rich grammar editor, coverage tools, phrase interpretation, seman- how-to’s and real-world examples drawn from the experiences of two large tics stepping, intelligent sentence generation, and so on. financial services companies.

Leveraging Existing Reporting to The Impact of W3C Standard Languages D106

Rapidly Tune Speech Applications M 4:15 p.m. – 5:00 p.m. David Claiborn, Senior Designer, United Healthcare O A common misconception in speech is that one can only tune an applica- MODERATOR: James Barnett, Senior Technical Director, Aspect Software tion with thousands of transcriptions. When standard transcribed data is The publication of W3C standard languages, such as VoiceXML and N not available, algorithms generate grammar weights based on standard CCXML, has dramatically changed the speech application design process. D reporting data. This presentation will cover techniques for using nontradi- This session discusses some of the efforts to extend and validate the use A Y tional data to deliver a rapid round of tuning. of standard languages. First, learn how the call control language can work ,

with SIP and VoIP to implement an extensible SIP softswitch. Then discover A U Integrating Speech Technologies how the VoiceXML Forum’s certification program has impacted the cross- G D105 vendor interoperability of VoiceXML by VoiceXML platform vendors.

With Enterprise Applications U

3:15 p.m. – 4:00 p.m. SIP Applications Using CCXML S MODERATOR: Rob Kassel, Vice President Marketing & Product, R.J. Auburn, Chief Technical Officer, Voxeo T

Holly Connects Developers are increasingly using the W3C Call Control language, CCXML, 1 Services-oriented architecture (SOA) will be a significant area of technol- to add call control features to their telephony applications. At the same time, 8 ogy investment for enterprises. New technologies make it possible for con- Session Initiation Protocol (SIP) and Voice over IP (VoIP) have found wide- tact centers to tap into the SOA to drive more advanced speech applica- spread acceptance and deployment in everything from long-distance net- tions with shorter development times and less risk. This session will focus works to enterprise call centers and in consumer telephony. Learn how SIP, on the technical standards and practices that allow speech-enabled IVRs VOIP, and CCXML can work together in next-generation telephony deploy- to benefit from existing and upcoming SOA investments. The session will ments, plus the advantages and disadvantages of a combination of both. also emphasize practical tips and real-world examples. Learn how CCXML can be used to implement an extensible SIP softswitch. Integrating Speech Applications With VoiceXML Platform Certification: Enterprise IT Assets Using Web Services What Every Customer Should Know Michael Codini, Chief Technical Officer & Co-Founder, VoiceObjects Ken Rehor, Founder, VoiceXML Forum One shortcoming of many speech applications deployed today is the lack VoiceXML platform certification has helped to stabilize and mature the of integration between these applications and the rest of an enterprise’s IT speech and contact center industry by encouraging the cross-vendor inter- infrastructures. As a result, valuable opportunities to enhance speech appli- operability of platforms, tools, and applications. Along the way, certification cations using intelligence contained in an organization’s web, CRM, and has become a key speech system RFP criterion for savvy companies adopt- other IT assets are lost, and IT efficiency suffers. web services integration ing speech. This presentation will discuss why VoiceXML platform certifi- between speech applications and other enterprise IT assets can remedy cation is an indispensable requirement of a complete speech system spec- this situation. Learn how web services can be applied to call center and ification. This presentation will also detail the pitfalls that customers may other speech-enabled environments and how to fully leverage CRM, BI, encounter in choosing self- or vendor-certified platforms. A summary of and other IT assets via web services. what is covered and what is not covered by the VoiceXML platform certifi- cation will be presented. How Speech Recognition Works in the Service-Oriented Architecture John Oh, Technical Lead, Customer Contact Business Unit, Cisco Sponsored by According to Gartner, spending on service-oriented architecture is expected Welcome Reception to grow from $14 billion in 2005 to $189 billion in 2009, making it the single-most significant area of technology investment for the enterprise. All SpeechTEK attendees are invited to a Grand Opening Reception New technologies make it possible for contact centers to tap into the SOA from 5:00 p.m. – 7:00 p.m. in the Exhibit Hall on Monday. to drive more advanced speech applications with shorter development times and less risk. This session will focus on the technical standards and Network with exhibitors, speakers, and conference attendees practices that allow a speech-enabled IVR to tap into the benefits of an exist- while enjoying a glass of wine and light hors d’oeuvres. ing or upcoming SOA investment. The presentation will include practical

8:00 a.m. – 8:50 a.m. Monday Sunrise Semester Discussions w SUNRISE SEMESTER DISCUSSION 2 Royale Room (6th floor) The Effectiveness of Marketing Messages, Legal Verbiage, and SUNRISE SEMESTER DISCUSSION 1 Website References in IVR Systems Emotion in Speech Technology DISCUSSION LEADER: DISCUSSION LEADER: Renae Rogers, Dialogue Designer, West Corp. Paolo Baggia, Director of International Standards, Loquendo SpA Marketing messages, legal verbiage, and web refer- Join us for a short overview of how emotions impact on speech applications, including ences in speech applications all compete for the time speech synthesis, speech recognition and facial expressions. Possible discussion ques- your customer has to listen and speak. Improper place- tions include the following: ment and verbiage can distract the caller and reduce • What are the types of emotion that would be useful to detect during speech recognition? the effectiveness of even the best speech application. • How would an application respond when it detects various emotions? Learn how customers feel about these messages and • How would an application express emotions? some practical lessons to maximize their affect while • How can existing applications be enhanced by emotions in speech technology? not hurting your prospects for voice self-help. • What new applications are enabled by emotions in speech technology?

WWW.SPEECHTEK.COM 17 Keynote SPONSORED KEYNOTE LUNCH w Sponsored by Remember Me: A Case Study on Personalized Self-Service 9 12:30 p.m. – 1:45 p.m. Broadway Ballroom, 6th floor 1 Preparing for the New Consumer w 9:00 a.m. – 10:00 a.m. Broadway Ballroom Tracy Griffith, Sr. Manager Customer Technology Reservations T w S Lior Arussy, author of several books, including his most recent, & Premium Services, American Airlines

U Excellence Every Day: Make the Daily Choice—Inspire Your In 2007, American Airlines launched an industry-leading telephone service

G Employees and Amaze Your Customers (Information Today, Inc.) for its frequent fliers to deliver proactive information based on the callers’

U Despite technological innovations, companies are failing to meet upcoming needs. A year later, find out how American Airlines did it and A increasing customer expectations. To stay competitive, organizations whether the investment has paid off — both for travelers and for American ,

Y must meet customers’ growing demands by incorporating new ways Airlines. AAdvantage members can enroll in “Remember Me” with the A of thinking, connecting, and doing business that will create profitable instructions at www.aa.com/callaa.

D and delightful customer experiences. Lior Arussy is a renowned S

E author; business visionary; creative catalyst; and president of Stra- A203 U tivity Group, a consultancy which advises Global 2000 companies Cross-Channel User Experience

T and emerging businesses around the world. Arussy will address the 1:45 p.m. – 2:30 p.m. biggest customer strategy issues facing organizations today, such MODERATOR: Eduardo Olvera, Senior User Interface Designer, as the new rules of customer engagement, and offer suggestions Nuance Communications on how to better prepare for the new consumer. The days of having a speech strategy separate from an overall user-experi- ence strategy are over. For a strategy to be meaningful, organizations must consider speech technology as one element in a multichannel customer con- TRACK A w Soho Complex (7th Floor) tact plan that encompasses all the ways in which the customer might inter- MEETING BUSINESS GOALS act with them. This session will provide valuable lessons in understanding WITH SPEECH TECHNOLOGY how speech fits with the web, printed material, live customer service, and other channels through which organizations touch their customers. Speech: Thinking Out of the Box A201 Is Your Caller Experience Web-Aware? 10:45 a.m. – 11:30 a.m. Lizanne Kaiser, Customer Experience Designer, Genesys MODERATOR: Judith Markowitz, President, J. Markowitz Consultants Telecommunictions Laboratories PANELISTS: The rapid expansion and penetration of the web has dramatically changed Whitney Quesenberry, Consultant/Researcher, WQ Usability customer behavior not only online but also during phone calls. Despite this, Moshe Yudkowsky, President, Disaggregate automated phone systems are often still designed as siloed channels. Steve Chirokas, Vice President of Marketing, SpeechCycle Speech systems that lack web-awareness provide incoherent customer What are the bleeding-edge possibilities for speech technology? In this experiences, damaging customer loyalty and business benefits. Drawing panel we take 3 steps back to reexamine speech from the outside to get a from client case studies, this presentation will explore the impact of the web new perspective on where we stand today and where we could be tomor- on customer behavior and expectations, metrics for assessing the web- row. Join us for a forward-thinking conversation on how speech can be an awareness of your caller experience, design guidelines for creating web- empowering technology for the automated, mobile interfaces of the future. aware speech systems, and predictions about the evolving convergence of the web and phone, along with the ramifications for speech technology. Video in the Network and Call Center A202 11:45 a.m. – 12:30 p.m. Speech and the Cross-Channel Customer Experience: MODERATOR: Valentine Matula, Director, Multimedia Research, Avaya Leveraging Speech to Improve Branding, Customer Is there a place for video alongside your speech solution? Experts in this Satisfaction, and Self-Service Rates session make the case for how video technology offers new opportunities Elaine Cascio, Vice President, Vanguard Communications in the call center and next-generation networks. Learn how video will change Customers conduct business with companies through many different chan- the ways we approach self-service and about the new business opportu- nels, but the experience is often disjointed and inconsistent. This creates nities video will enable. customer confusion and frustration or, even worse, abandonment for the competition. They’ll also tell at least five others about their dissatisfaction. Video-Based Call Centers and Self Service Applications: In this session, you’ll learn how your speech application is a critical com- Is There an Untouched Need Out There? ponent to building a positive cross-channel customer experience. We will Bob Cooper, Chief Architect, Voice Portal, Avaya examine how speech can contribute to the overall multimedia brand expe- How will the customer experience change when video-based call centers rience and build customer loyalty, plus we’ll look at real-life examples of and self-service applications become available? What new services that we organizations that provide seamless cross-channel service (as well as some are not considering today will become commonplace and what will they that don’t). Finally, we will outline the critical steps you should take to ensure mean to your call center? This session will look into a few scenarios and your customer has an engaging, seamless, and positive experience. discuss the technical hurdles that must be overcome. You will also learn which international locales are best prepared to deploy these technologies. Speech in CPG/Retail A204 Opportunities for Video in the Next-Generation Network 2:45 p.m. – 3:30 p.m. Rob Marchand, Senior Director, Product Management, Genesys MODERATOR: Sunil Issar, Director, Convergys Telecommunications Laboratories Consumer packaged goods companies and retailers rely on speech tech- This session will discuss opportunities for the deployment of speech- nology in a variety of ways. This session will focus on two of them: ware- enabled video applications in the enhanced-services and managed-serv- housing and customer self-service solutions. One presentation will feature ice environments. The discussion will include overviews of real-world Coca-Cola’s warehouse voice picking system, based on inexpensive IP- deployments, plus a review of the challenges that remain. Topics will cover phones and network-based speech recognition, which yielded significant architecture, applications, and standards, including SIP and VoiceXML. cost reductions.

18 AUGUST 18–20, 2008 w NEW YORK MARRIOTT MARQUIS, NEW YORK CITY TUESDAY, AUGUST 19

VoIP Creates New Approach to Speech-Based Warehousing leads to direct business benefits, including faster calls, happier customers, productive agents, and lower phone bills. Yes, you can have your cake and CCS Michael Jacks, Senior Manager, Logistics & Transportation, Coca-Cola Enterprises eat it, too. Coca-Cola Enterprises created the first-ever deployment of warehouse voice-picking based on inexpensive IP phones and network-based speech ATTENDEE RECEPTION w On the 9th Floor T recognition. The results? Significant reductions in per-device costs and total 5:30 p.m. – 7:00 p.m. U E

cost of ownership. This presentation will provide insights into what drove S this innovation, the results achieved, and future plans to leverage speech D capabilities on a services-oriented architecture (SOA) basis throughout the TRACK B w Empire Complex (7th Floor) A company. VUI DESIGN PRINCIPLES AND TECHNIQUES Y , B201 A Guided Self-Service Solution for Call Centers How Do You VUI? U

CCS Tom Scott, CIO and SVP, Operations, Spiegel Brands 10:45 a.m. – 11:30 a.m. G

The Spiegel call center has replaced the traditional menu-driven touchtone MODERATOR: Peter Krogh, Speech Mechanic, SpeechCycle U approach to receiving and routing calls with guided self-service. This ses- There’s more than one way to design a good voice user interface. Join S T sion takes an up-close look at the new solution, which enables a behind- us for a session in which VUI experts give us new ways of thinking about the-scenes human guide to monitor and assist four or more simultaneous VUI issues. Learn how the precise wording of VUI prompts can vastly influ- 1 9 calls, resulting in a better caller experience with reduced ordering and cus- ence the way users respond and for a discussion on the impact new VUI tomer service costs. standards will have on our industry. Come and participate in what prom- ises to be a spirited debate of some of the most important issues in VUI The VUI Backlash A205 design today. 4:15 p.m. – 5:15 p.m. Coffee? Tea? Yes, Please MODERATOR: Lizanne Kaiser, Customer Experience Designer, Genesys David Suendermann, Speech Science Engineer, SpeechCycle Telecommunications Laboratories Ethan Levine, Voice Interaction Designer, SpeechCycle Touchtone IVRs was once the technology people loved to hate—an object When designing a troubleshooting application, much effort is put into find- of ridicule in popular culture, one of the slings and arrows of modernity we ing an optimal balance between eliciting in-grammar responses and brevity were forced to suffer. Speech recognition was billed as a solution to the ills of dialogue. This session will present data from calls to live systems that of touchtone, but there are rumblings that speech-enabled IVRs are rapidly used three styles for eliciting responses: giving the caller explicit instruc- taking the place of touchtone in the cultural hall of shame. In this session, tions as to the set of valid responses; phrasing a question to include the we examine what the general public really thinks about speech and exam- set of valid responses, though not explicitly stating how the caller should ine why speech hasn’t lived up to the initial hype. The speakers will explore respond; and phrasing a question without any guidance. Of note, certain the reasons people are dissatisfied and ways to alleviate their distress. formulations encouraged ambiguous responses; we will discuss relevant Why People Really Hate IVRs and What to Do About It avoidance and recovery strategies for these. Ahmed Bouzid, Senior Product Manager, Angel.com Getting Serious About IVR Dialogue Standards: IVR systems are generally perceived by users as obstacles installed by com- A Framework for Action panies to keep callers from reaching expensive human agents, rather than Bruce Balentine, EVP and Chief Scientist, Enterprise Integration Group helpful tools that can effectively serve callers’ needs. IVRs are not only fail- Ken Rehor, Founder, VoiceXML Forum ing to do their jobs, but they are also pushing some of users’ most sensi- User interface standards offer both utility and disruption to the many play- tive hot buttons. This session will identify these emotional triggers and offer ers involved in IVR, including end users, enterprise stakeholders, call cen- some clear guidelines on ensuring the deployment of highly usable IVR ter agents, technology vendors, and VUI designers. Conflicts of interest and solutions. philosophy constitute a culture of resistance that has reduced IVR quality Get Human, Get Real and raised costs worldwide. The goal of this presentation is to propose a Simonie Wilson, Senior Speech Scientist, Intervoice framework for which an organized and representative community of stake- Some key issues have been missed in the ongoing discussion about GetH- holders within the IVR industry can go about agreeing on a standardized uman, a movement founded by Paul English and based on his belief that set of IVR behaviors for telephone-based dialogues. a live agent is always better than an IVR of any kind. The truth is, it’s not that simple. From a desire for privacy, to needing assistance in the middle Inaugural Meeting of the Association of the night, to a dislike for offshoring, callers have many reasons to prefer for Voice Interaction Design B202 automated systems instead of live agents. We’ll discuss how to improve the 11:45 a.m. – 12:30 p.m. automated part of a call, and maybe we can offer some suggestions for the MODERATOR: Susan Hura, Principal, SpeechUsability GetHuman initiative. At SpeechTEK 2007, a group of user interface designers met and agreed Have Your Cake and Eat It Too that our profession needs an organization to promote excellence in voice Nick Ezzo, Director of Marketing, TuVox interaction design, facilitate professional development, and provide an opportunity to share experiences with other designers. The result of that Many people believe that providing an excellent caller experience is an 2007 meeting is the Association for Voice Interaction Design (AVID). Join unnecessary expense—that “caller experience” and “business benefits” are us for the inaugural meeting of AVID for the opportunity to vote on the pro- contradictory terms. Based on data from actual implementations, this ses- posed charter and elect officers. Read the charter online at www.avixd.org sion will methodically build a case that proves a great caller experience before the meeting and come prepared with questions and suggestions.

Breakfast & Coffee Breaks SPONSORED KEYNOTE LUNCH w Sponsored by Remember Me: A Case Study on A continental breakfast will be provided for conference attendees Personalized Self-Service each morning from 8:00 a.m. to 9:00 a.m. before the keynote session. 12:30 p.m. – 1:45 p.m. w Broadway Ballroom, 6th floor Additional coffee breaks will take place each morning and afternoon. Tracy Griffith, Sr. Manager Customer Technology Reservations Please check the schedule for exact times and locations. & Premium Services, American Airlines WWW.SPEECHTEK.COM 19 lobbyists, etc. This session features a demonstration of Prime III, as well as B203 Business Problems, VUI Answers a discussion about how speech was incorporated into this award-winning 1:45 p.m. – 2:30 p.m. application, which could change the way we vote. MODERATOR: David Martin, Principal Consultant-Speech Engineering, Avaya Mark Webb, Process Engineer-IVR, Humana Talking Games: Toward Speech as a Mainstream Modality 9 David Thornton, Instructor, Auburn University 1 Daniel Padgett, Sr. VUI Consultant, VoicePartners This session will describe the results of an ongoing study of speech-based

T Come to this session to hear the inside story of the VUI process from two

S perspectives. In the first presentation, the customer organization, Humana, cursor control mechanisms—research that is intended to provide user data

U presents the business challenges that led the company to seek a speech to influence the design of future systems involving real-time demands, very

G solution. Next, the solution provider, VoxGen, presents its view of the situ- small targets, and moving targets. One such application of this research is

U ation and discusses the methodology used to develop a highly effective in the area of video games. These findings also have implications for phys- A solution for its client. Join us for this unique opportunity to talk with both ically impaired users whose primary or only control modality is speech. ,

Y the customer and the vendor together to understand how to work as a team

A to solve business problems with speech. Driving the Conversation

D B205

S Forward in the Face of Errors B204 E VUI Design for Multimodal Applications 4:15 p.m. – 5:15 p.m.

U 2:45 p.m. – 3:30 p.m. MODERATOR: Jim Milroy, Director, Creative Services, West Interactive T MODERATOR: Jonathan Bloom, Senior Interaction Designer, SpeechCycle Users encounter error conditions in every voice user interface, and VUI Here you’ll see multimodal user interface design in action. These presen- designers must plan for how they will deal with inevitable errors. Conversa- tations will give real-life examples of multimodal user interfaces incorporat- tional errors occur regularly in human-to-human communication as well, but ing speech technologies. Presenters will highlight the challenges of inte- we deal with them so seamlessly that the positive flow of conversation is grating speech with a visual interface and describe the techniques they rarely interrupted. The way we deal with conversational errors in VUIs is rarely used to design effective, pleasing user interfaces. Come to see demonstra- this smooth and often begins a needless progression toward the failure of tions of both a multimodal voting system and multimodal games with the interaction. The experts in this session explore new ways to look at error speech controls and decide if multimodal is ready for prime time. conditions and methods for dealing with errors, with the goal of building VUIs that de-emphasize errors and move conversation forward. Prime III: A Multimodal Approach to Electronic Voting Juan Gilbert, T-SYS Distinguished Associate Professor, Auburn University A New Perspective on Speech Recognition Errors Prime III is a multimodal electronic voting system designed, implemented, Gregory Simsar, Vice President Speech Services, Syntellect and tested by the Human Centered Computing Lab at Auburn University. Many types of speech communication behaviors that humans handle effort- The system has caught the attention of Congress, local and national media, lessly are not even detected by today’s speech recognition systems, let

Making voice self-service easy Stand 306

Vicorp can make it easy for your organization to cost-effectively deliver superior automated self-service experiences over the phone How do we make it easy for you?  Our technology: Smart, easily customized and based on open industry standards

 Our people: Experts ready to inspire your organisation

 Our accountability: Our solutions deliver measurable business benefit

 Our approach: We work with our customers to understand their callers' needs, business challenges and brand identity

For more information come to Stand 306 www.vicorp.com [email protected] +44 1753 660 500

20 AUGUST 18–20, 2008 w NEW YORK MARRIOTT MARQUIS, NEW YORK CITY TUESDAY, AUGUST 19 alone handled gracefully. Only by taking a new perspective on the types of ity checked by human interaction or runs unattended. Key ideas include errors that are encountered by speech recognition systems—and by map- the automatic adaptation of the language model to the interests of the user ping them to the types of errors that occur in human-to-human interaction— which allows deployed systems to adapt to dynamically and quickly chang- can we start to make significant improvements in how errors are handled. ing environments without interaction by the vendor of the system, as well

This presentation builds on a SpeechTEK VUI workshop and subsequent as identification of the ideal point in time to rebuild a language model. T article in Speech Technology magazine about a new taxonomy for speech U recognition errors. Advancing Toward Intelligent Agents E Emmett Coin, Speech Scientist, ejTalk S D The Subtle Science of Failing: How to Retry Voice applications began as almost pure speech recognition systems: Jonathan Bloom, Senior Interaction Designer, SpeechCycle A speaking choices, numbers, etc. Primitive directed dialogues followed. Now Y Jackson Liscombe, Speech Engineer, SpeechCycle many applications support some sort of mixed initiative as they progress , Just like humans, speech recognition systems do not always understand A

toward less-rigid behaviors. Sadly, these methods are already failing to meet U their conversational partners. How should these systems recover when

the expectations of users, not to mention leading to paralyzing complexity G recognition confidence is low? Up until now, designers have based their

for developers. Going forward, voice systems will incorporate other exist- U retry strategy mostly on tradition, bombastic opinion, anecdotal evidence,

ing and proven technologies to make applications smarter. These technolo- S or, at best, one-off opinions from usability test participants. In this presen- gies have the potential to dramatically reduce the complexity of develop- T tation, we will discuss research based on tens of thousands of calls, each ment for today’s sophisticated voice applications. This session will compare 1 hitting one of four different retry styles: an apologetic preamble, a pream- 9 and contrast the benefits and limitations of these technologies ranging ble without an apology, an abbreviated retry, or a retry avoiding explicit ref- across natural language processing, rule-based and case-based reason- erences to self. We will base the relative success of each strategy on turn ing, and statistical and connectionist (neural net) approaches. success rate, average number of retries required per turn, percentage of out-of-grammar utterances, automation rates, and caller satisfaction. Enhanced VUI: Context-Sensitive Dialogue C202 Refocusing Caller Intent: Approaches for Building 11:45 a.m. – 12:30 p.m. Fault-Tolerant Voice User Interfaces MODERATOR: Tom Schalk, Vice President, Voice Technology, ATX Jessica Peterson, Speech Technology Consultant, and Natural language understanding (NLU) technologies can significantly Based on real-world deployments, this session will present detailed improve caller experience and improve business metrics such as routing aspects of a framework for natural language routing applications. You will learn how to adapt confirmation states to specific types of initial accuracy and call completion rates. The ultimate goal is to deliver intelli- caller requests to elicit usable input that leads to successful call routing, gent responses to natural speech input. Based on the current state of as well as strategies for handling callers who provide input to a natural speech recognition technologies, NLU technologies need to be used selec- language main menu that cannot be successfully mapped to a business tively to achieve the best performance. For example, it may be better to use category. You’ll gain a new approach that creates a second chance to a menu followed by an open-ended prompt instead of the reverse. This ses- harness the benefit of natural language recognition when warranted, sion dives deep into NLU technologies and how to build robust statistical with the goal of creating a smarter speech application and improving the models for optimal performance. Interesting application performance overall user experience. results are discussed, including effective handling of spontaneous speech. ATTENDEE RECEPTION w On the 9th Floor Building Robust NLU Applications 5:30 p.m. – 7:00 p.m. Sunil Issar, Director, Convergys Natural language understanding (NLU) technologies can significantly improve caller experience and improve business metrics such as routing TRACK C w Majestic Complex (6th Floor) accuracy and call completion rates. These technologies allow the caller to ADVANCED SPEECH speak naturally. In general, based on the current state of speech recogni- TECHNOLOGIES SYMPOSIUM tion technologies, NLU technologies need to be used selectively to achieve the best performance. For example, it may be better to use a menu followed Natural Language Processing: Automaticity C201 by an open-ended prompt. We will describe when it is appropriate to use 10:45 a.m. – 11:30 a.m. NLU technologies and how to build robust statistical models for optimal performance. We will also describe results comparing performance of MODERATOR: Bill Scholz, President, NewSpeech LLC directed dialogue and statistical grammars. Key Ideas: NLU can improve The use of statistical natural language processing continues to expand, but business metrics; NLU technologies need to be used selectively for opti- its growth pace is inhibited by the cost of language model development mal performance; VUI design and statistical models need to handle spon- and maintenance. This session describes a framework that automatically adapts and changes automatic speech recognition language models over taneous speech, which introduces additional complexities. time and considers the evolution of applications from pure speech recog- Intelligent Responses to User Input nitions systems, to mixed-initiative systems with less rigid behavior, and to Susan Boyce, Principal Designer, Tellme (a subsidiary of Microsoft) voice systems that incorporate proven technologies to become not only Customizing voice applications for individual users can have several ben- smarter but less complex to maintain. eficial effects. Accurately anticipating the reason for a call can streamline CAVA: Continuous Automatic Vocabulary Adaptation the interaction. Other characteristics of the caller, such as location or pre- Mark Pfeiffer, VP Business Development & Communications, SAIL LABS vious calling history, can be used to enhance recognition performance. Technology AG This presentation will review examples of effective use of personalization CAVA from SAIL LABS is a framework that automatically adapts and in voice applications and set forth some guidelines. changes automatic speech recognition language models over time to cur- rent affairs. The user determines areas of interest that are to be harvested SPONSORED KEYNOTE LUNCH w Sponsored by for domain- or topic-specific data, which subsequently form the basis of automatic language model creation. Data sources are combined accord- Remember Me: A Case Study on ing to their relevance to yield a series of domain-, time-. and topic-specific Personalized Self-Service language models. This mix of language model data and sources then adds 12:30 p.m. – 1:45 p.m. w Broadway Ballroom, 6th floor and subtracts words and terms depending on their relevance in the cur- Tracy Griffith, Sr. Manager Customer Technology Reservations rent deployment environment. This processed data is then optionally qual- & Premium Services, American Airlines

WWW.SPEECHTEK.COM 21 22 TUESDAY, AUGUST 19 a e CRM vidual caller'svalue, the current state ofcallcenter queues, information from IVR applicationsinthat itutilizesablended strategy anindi- that considers t able resource (either automated orlive agent) to effectivelyresolve most onhowsion basedonbusinessrules to route the callerto the suit- most textual information from their profile andhistory, andthen makes adeci- The iCFDno Phil Shinn, The IntelligentCustomerF text ofthe callcanhelpfacilitate numerous intelligent adaptations. Beingaware ofthe con- tiation context, andcallpopulationdistribution. agement ofaninteraction, such asuser profile, recent callerhistory, callini- session willcover several typesofcontexts that canintelligently inform man- This in the interaction andwhatexactly the system expects you to donext. to itagain;andinvisible—you can’t figure easily outwhere precisely you are directional—when you hearsomething you can’t goback easily andlisten listenpatiently to oneword before you canhearthe onethat follows it; uni- VUIs are restrictive inthree ways: crucial They are timelinear—you must Ahmed Bouzid, The LimitsandPossibilities ofVUIDesign S this growing inuserinteraction sophistication usingalarge subjectpool. behavior andwillreview experiments designedto evaluate the benefitsof s todynamically the overall context ofthe conversation. Speakers inthis ses- r As speech applicationsgrow they insophistication, are moving beyond MODERATOR: 1:45 p.m.–2:30 E Genesys Telecommunications Laboratories AUGUST 18–20, 2008 he interaction. Thisdynamicapproach isdifferent from traditionalstatic igid preconceived dialogueflow toward userinteraction that responds ion considerrecent success inbuildingapplicationswith this evolving xper hifting theIntelligenceBurden: nhanced VUI:DynamicResponse Generation iCFD, anddiscusssomecasestudies. Nortel_HalfH_STEK08_FinalPrg 7/15/083:09PMPage1 ience. Thist databases todatabases on-the-fly, construct, caller aunique andpersonalized Principal Consultant,Principal Voice ApplicationServices, t onl Bill Scholz, Senior Product Manager, Angel.com fee modelthat y alk wit greets acaller, their butalsogathers intent, addscon- Nortel’s turnkey Hosted solution IVR offers astraight-forward, per-minute TRUMPS SIMPLICITY h numbers, transport, superiorspeechapplications, and24x7 support. focus onthe methods andfeatures usedto build President, NewSpeech LLC ront Door(iCFD) BUSINESS MADESIMPLE. includes everything neededfor speech IVRsuccess CO w E OKMRIT MARQUIS MARRIOTT YORK NEW MPLEXITY nortel.com/hostedivr C203 ent used atspecificplacesinadialogue. tions andwit request. Excruciating menusare avoided by direct accessto func- most ing technology, andfree-form recognition ofmultipleitems inasingle language modeling,naturallanguage understand- statistical state-of-the-art anddialoguemanagement featuresguage understanding include tems to accomplish their objectives andefficiently. easily Key naturallan- with naturally verse devices such asmusicplayers andGPSnavigation sys- technologies. Examples ofthis are systems to that willenableusers con- inembeddednaturallanguage interactionest anddialoguemanagement This presentation willfocus onautomotive that interfaces illustrate the lat- Sicconi, Roberto Conversing WithYour Car in what’s upandcominginautomotive speech. for conditions.Thissessionisamust under noisydriving those interested modelstargetedtion andspecializedacoustic athandlingfar-field speech foritself, alsocritical speech enablementisproper microphone configura- center, andhow itcanbeappliedinthe vehicle. Beyond the userinterface interaction anddialoguemanagement technologies from amajor research Thissessionwillrevealdriving. the latest inembeddednaturallanguage a toallows drivers usespeech to control music,phones,navigation systems, T MODERATOR: 2:45 p.m.–3:30 A nicel The automobile isdesignedto besafe to fit drive, andspeech interfaces Tom Schalk, Voice AutomationintheVehicle: Deployments andTrends tionality use speech to control music,phones,navigation systems, andother func- nd other functionalities,making itpossibleto bemore productive while he human/machine interface inthehe human/machine interface vehicle isevolving quickly andnow dvances inAutomotiveSpeech ation willanal y int , making itpossiblet o t he driving experience.he driving Today, to allow speech drivers interfaces hout having to remember specificcommandsthat needto be Vice President,Voice Technology, ATX , Tom Schalk, E OKCITY YORK NEW Program Director,Program T.J. Watson Research, IBM yze significantaut Vice President,Voice Technology, ATX o be more productive Thispres- whiledriving. omotive speech deployments and — telephone C204 TUESDAY, AUGUST 19 include topics such as microphone configuration, acoustic models, lan- showing advanced personalization features, ANI authentication, and context- guage models, and multimodal interface requirements. specific menus so callers are offered appropriate choices at any point.

Ancillary Transcription Techniques C205 Optimizing a Caller’s Request for a Live Agent Patrick Nguyen, Chief Technology Officer, Voxify 4:15 p.m. – 5:15 p.m. T

How your self-service application handles that desperate cry for an agent U MODERATOR: Bill Scholz, President, NewSpeech LLC

is a careful balancing act. The difficulty lies in conflicting objectives—con- E

Recognition technology has matured to the point that recorded telephone- tact centers want to help callers facing problems but avoid unnecessary S quality audio from unknown speakers can be transcribed with sufficient transfers. If the application transfers without analyzing the situation, the D accuracy to support commercial applications such as medical data tran- A

caller might face long hold times and more frustration. New research on Y

scription, broadcast news captioning, and near-real-time conversion of ,

the contextual handling of transfer requests have resulted in a flexible voicemail to text. Recent improvements in transcription quality are achieved approach for dialogue design where the speech application facilitates rather A U through the use of ancillary information such as caller-specific data serv- than impedes the completion of callers’ goals. ing to narrow that speaker’s domain, speaker-dependent acoustic models G from frequent callers, and the use of CFGs to identify words and phrases “Welcome Back, Steve,” Said the Friendly ... Computer? U that occur with high frequency in the target domain. This session will review Steve Pollock, EVP and Co-Founder, TuVox S T

the most beneficial ancillary factors contributing to transcription accuracy It’s amazing how quickly you get used to personalized customer service. and illustrate the effects of their incorporation. 1 Most consumer websites offer a personalized experience that is highly tai- 9 lored with readily accessible account information. Can this happen over the Ancillary Transcription Techniques phone? This session will describe a recent application at a major Ameri- Michael Picheny, Manager, IBM Research can airline with advanced personalization features, including ANI authen- As applications of large vocabulary speech recognition proliferate, there tication, greet-by-name, and context-specific menus so that callers are are increasing opportunities to use various types of side information to offered appropriate choices at any point. enhance recognition accuracy. This presentation will explore a range of side information, ranging from the talker’s identity to more subtle informa- Sponsored by tion from extrinsic sources, such as questionnaire responses and cross- SPONSORED KEYNOTE LUNCH w language constraints. The talk will also address the issues in incorporating Remember Me: A Case Study on various additional information sources in practical deployable systems. Personalized Self-Service Speech and Speaker Recognition 12:30 p.m. – 1:45 p.m. w Broadway Ballroom, 6th floor Systems Can Make Use of More Information Tracy Griffith, Sr. Manager Customer Technology Reservations Jordan Cohen, Senior Scientist, SRI & Premium Services, American Airlines Speech and speaker recognition systems can make use of more informa- tion than is normally included in an acoustic model. We have been exper- Speech in Healthcare D203 imenting with the use of back-channel information, word choice, channel 1:45 p.m. – 2:30 p.m. modeling, and other measures to assist in performing better speech recog- MODERATOR: Raj Tumuluri, Principal, Openstream nition. In addition, we have started a project to collect data in which speaker With an ever-increasing focus on accuracy and privacy in the healthcare style is explicitly controlled. Appropriately modeling the aspects of style is industry, speech technology is playing a bigger role. The first case study expected to enhance our recognition performance substantially. in this session shows how a natural language VUI enables bidirectional access to the entire text-based component of electronic medical records. ATTENDEE RECEPTION w On the 9th Floor Learn how this application was designed, implemented, and deployed and 5:30 p.m. – 7:00 p.m. how it impacts medical center operations. The other presentation highlights how healthcare companies can use speech technology for outbound calls. Current healthcare protocols often require regular patient follow-up after a TRACK D w Shubert Complex (6th Floor) hospital discharge or as part of a clinical study. ATS found an IVR applica- DEVELOPMENT AND DEPLOYMENT tion that is just as effective as live nurse calls, more cost effective, and more beneficial—providing more consistent, reliable calls. D201 Securing CCXML and VoiceXML Applications ICIPS: Integrated Clinical Information Phone Service 10:45 a.m. – 11:30 a.m. CCS Val Nenov, Adjunct Professor, Division of Neurosurgery, MODERATOR: Jim Larson, VP, Larson Technical Services UCLA Medical Center Dan York, Director of Emerging Communication Technology, Voxeo ICIPS is an ongoing R&D project at the UCLA Medical Center. Its main objec- How secure are your speech applications? As the usage of both VoiceXML tive is to provide bidirectional access to the entire text-based component of and CCXML continues to explode, and VoIP usage continues to grow dra- the electronic medical records using a natural language VUI. Learn how this matically, especially within enterprise environments, it is increasingly impor- application was designed, implemented, and deployed and how it impacts tant that you ensure that applications and services are not open to attack. the operation of a medical center. This presentation will summarize the Learn about the potential vulnerabilities in a system using VoiceXML or strengths and weaknesses of the VUI as compared to an existing GUI. CCXML, what you can do to secure these systems, and how you can develop a strong architecture. When Is a Virtual Nurse Call Better Than a Live One? CCS David G. Engelhardt, President, Advanced Technical Support, Inc. D202 Personalization and Context Current healthcare protocols often require regular patient follow-up after a 11:45 a.m. – 12:30 p.m. hospital discharge or as part of a clinical study. Typically, doctors rely on MODERATOR: R.J. Auburn, Chief Technical Officer, Voxeo nurses to call patients to monitor recovery or to gather data for statistical Many consumer websites offer a personalized experience that is highly tailored analysis. These calls are expensive and time consuming compared to an with readily accessible account information. But what about over the phone? IVR application that can replace nurse calls. Automated calls were found This session will review approaches to using context to improve the user inter- to be just as effective as live nurse calls and had additional benefits—more face dialogue. The first presentation describes recent research on the contex- consistent, reliable calls were possible. This presentation will discuss the tual handling of transfer requests. The second presentation describes recent study protocol and the results of the trial. Virtual nurse calls now are a applications for the airline, entertainment, and financial services industries, proven, winning strategy for clinical studies and patient care.

WWW.SPEECHTEK.COM 23 TUESDAY, AUGUST 19

based on two metrics: number of callers successfully matching their need D204 Speech in Government/Utilities to a technical support topic and automation rate. One of the designs was 2:45 p.m. – 3:30 p.m. a clear winner, improving both accuracy and automation by 10%. The sec- MODERATOR: David Thomson, CTO, SpeechPhone ond case study, by Leaco Rural Telephone Cooperative, will cover how to

9 A clear sign that speech has hit the mainstream is when late technology improve the customer experience, brand differentiation, and identity fraud.

1 The third case study will provide an overview of a successful large-scale, adopters, such as governments and utility companies, are implementing

T speech solutions. The first case study in this session will describe the devel- customer-facing deployment of voice biometrics technology. S opment and deployment of a new speech application to provide customer U self-service to its San Francisco residents. The second case study dis- A Champion-Challenger Experiment for Designing a Natural G cusses why NSTAR chose speech to replace the previous menu structure, Language Question U integrating with other call center technologies, the importance of load and CCS Manohar Kesireddy, Business Solutions Architect, A

performance testing and customer reaction. Verizon Data Services ,

Y Verizon FiOS TV wanted to determine the best design for a natural lan- A Developing Call Center Applications guage question and back-off menu for a technical support application. The D CCS Marge Vizcarra, CSM, City and County of San Francisco company created three different approaches to challenge the then-existing S James Whitten, Assistant Manager, Contact Center, San Francisco Public E design. All four were deployed in a limited production release and evalu- Utilities Commission U ated based on two metrics: number of callers successfully matching their

T The San Francisco Public Utilities Commission will describe the develop- need to a technical support topic and automation rate. One of the designs ment and deployment of the new speech application to provide customer was a clear winner, improving both accuracy and automation by 10%. The self-service to its San Francisco customers. The goals of the new system characteristics of the winning design will be presented. are to increase use of automated offerings, free up agents to assist callers that can only be assisted by a live agent, save agent transfer calls for cus- Improve the Customer Experience With Voice Authentication tomers really needing assistance with an easy tool for self-service, and to CCS Laura Phipps, General Manager and EVP, Leaco Rural Telephone ensure that a live person is always reachable. Cooperative Leaco is a telecommunications services provider that supplies wireless, Improving Customer Experience Implementing wireline, and internet services to approximately 175,000 bilingual and mono- a Speech-Enabled IVR at NSTAR lingual customers in Southeastern New Mexico. In November 2007, Leaco CCS Michael Roberts, Business Integration Manager, Customer Interaction began rolling out customer-facing speaker-verification services to 12,000 Center, NSTAR Electric and Gas of its customers. The system is designed to address the following issues: In this presentation, attendees will learn why NSTAR chose speech tech- improve the customer experience; regulatory compliance (with CPNI); Cus- nology to replace the existing menu structure, how the interoperability with tomer concerns about security; and brand differentiation. other call center technologies affected the implementation, how load and performance testing ensured a successful implementation, and what reac- At Bell, My Voice Is My Password tion customers had to the new system. NSTAR’s telecommunications team CCS Fred MacKenzie, Sr. Business Solutions Advisor, Bell Canada learned many lessons during the rollout of the project on both a business Charles Giordano, Associate Director, Bell Canada and technology levels. Getting management’s commitment early in the Bell Canada is the largest telecommunications services provider in Canada. process and reviewing existing technology for any interoperability issues It has rolled out a nationwide, customer-facing system, starting with residen- are two of the lessons learned that will be shared during this presentation. tial wireline and wireless customers. By the start of 2008, Bell Canada had enrolled 500,000 customers. The company chose speaker verification for the Speech in Telecommunications D205 following reasons: to secure the privacy of customer data; to make privacy 4:15 p.m. – 5:15 p.m. more convenient; and to reduce the average length of call-center calls. MODERATOR: Judith Markowitz, President, J. Markowitz Consultants The first case study, by Verizon FiOS TV, will reveal three different ATTENDEE RECEPTION w On the 9th Floor approaches to challenge its existing design. All designs were evaluated 5:30 p.m. – 7:00 p.m.

Tuesday Sunrise Semester Discussions w 8:00 a.m. – 8:50 a.m. SUNRISE SEMESTER DISCUSSION 5 Royale Room (6th floor) W3C Multimodal Working Group DISCUSSION LEADER: Deborah Dahl, Principal, SUNRISE SEMESTER DISCUSSION 3 SUNRISE SEMESTER DISCUSSION 4 Conversational Technologies Speaker Identification It Ain’t Shakespeare ... Or Is It? Increasingly powerful mobile devices, along and Verification DISCUSSION LEADERS: with improvements in displays and speech DISCUSSION LEADER: Alexandra Auckland, Senior VUI Designer, Nuance technologies, are making possible innova- Judith Markowitz, President, Communications tive, compelling, and robust applications that J. Markowitz, Consultants Melanie Polkosky, Human Factors Psychologist and allow users to combine speech, graphics, An open meeting of the VoiceXML Forum’s Consultant, IBM ink, and motion. Emerging standards for Speaker Biometrics Committee, which is Like playwrights, VUI designers write dialogue that multimodal interaction support and simplify working on extending VoiceXML to include is meant to be spoken out loud. Yet, recent research the development and deployment of these Speaker Identification and Verification (SIV). reveals that less than 10% of designers in the field applications. Please join us for a discussion Anyone interested in SIV is welcome to have an educational background in voice perform- of current and planned work in the World attend. Discussion topics will include an ance, acting, coaching, voice pathology, or peda- Wide Web Consortium’s Multimodal Interac- update from the VoiceXML Forum’s Biomet- gogy. In this lively session, we will share techniques tion Working Group and learn how to get ric committee followed by a Q&A. about what to look for in auditioning talent, how to involved in the W3C standards process. write dialogue that translates well to the spoken lan- guage, and how to effectively coach a voice actor so the resulting application becomes more effective, natural, and usable.

24 AUGUST 18–20, 2008 w NEW YORK MARRIOTT MARQUIS, NEW YORK CITY AD enue. The session will identify caller experience information that organiza- Keynote tions must take into account.

0 DTMF Versus Speech A302 2 11:45 a.m. – 12:30 p.m.

T The Art of Deception: Are You

S MODERATOR: Phil Shinn, Principal Consultant, Voice Application Services, in Danger of Being ‘Conned’? Genesys Telecommunication Laboratories U 8:45 a.m. – 10:00 a.m. Broadway Ballroom G w How should an organization decide between speech and touchtone IVR solu-

U Kevin Mitnick, author, world-famous (former) hacker, & tions? The thinking used to be that speech was an all-or-nothing decision,

A security consultant but increasingly, organizations are discovering that a speech-plus-touchtone

, As more businesses and consumers depend on speech technology, solution may be the best fit for their IVR. Learn how to decide which features Y

A it’s becoming increasingly important to protect your investments. With require speech, which may be well-served by touchtone, and how to create

D more than 15 years of experience in exploring computer security, a smooth, cohesive user interface that includes both modalities. S Kevin Mitnick is a largely self-taught expert in exposing the vulnera- E bilities of complex operating systems and telecommunications Does Your Application Require Speech, Touchtone, or Both? N devices. As the world’s most famous (former) hacker, Mitnick has Darla Tucker, Principal, Global Consulting Services, Intervoice D been the subject of countless news and magazine articles published Buyers, beware: You might not need a big, expensive speech application E throughout the world. Mitnick will share how hackers compromise when a simple touch-tone application could be the perfect fit. Or perhaps W your security and what you can do about it. you do need the new speech application, but with a twist. No matter, the perfect always requires a solid case and strategy for what’s purchased and deployed. We’ll explore what you must consider and how to determine what best suits the needs of your business and your callers. TRACK A w Soho Complex (7th Floor) MEETING BUSINESS GOALS Balancing Business Needs Versus Caller WITH SPEECH TECHNOLOGY Needs When Migrating DTMF Applications Matt Whipple, Principle Consultant, Self-Service Solutions, Avaya Customer Satisfaction 2008 A301 DTMF applications are usually easier, faster, and cheaper to develop than 10:45 a.m. – 11:30 a.m. speech applications. As waves of proprietary IVR systems and hardware MODERATOR: Tim Moynihan, Vice President, Global Marketing & Sales platforms reach their end of life, they will need to be replaced. As a result, Support, Envox many businesses turn to vendors to replace those systems in a like-for-like Today’s consumers are increasingly sophisticated in their expectations for fashion—an approach that might saves time and money, but does it enhance customer service from the organizations with which they do business and customer loyalty and represent a service that meets caller needs? This pres- are increasingly willing to publicize negative experiences. This combina- entation will explore a few DTMF-to-DTMF application rewrites and how the tion leads to a volatile marketplace in which organizations must protect same business needs could have been better addressed with speech. their customers against poor user experiences to establish and maintain their competitive advantage. Experts in this session discuss how to real- LUNCH BREAK istically assess customer satisfaction and deliver a positive experience to 12:30 p.m. – 1:45 p.m. every customer every time. A303 You Can’t Hide From a Bad Experience Security for Speech Lynda Kate Smith, VP, General Manager, Core Business Unit, 1:45 p.m. – 2:30 p.m. Nuance Communications MODERATOR: Dan York, Director of Emerging Communication Technology, Organizations can no longer hide from their customers’ bad experiences. Voxeo Expectations are higher than ever before, and with today’s social media, How secure is your speech application? Organizations are now accustomed the consumer’s voice is louder and travels further. That’s why you must to dealing with security threats that arrive via email and web channels but ensure customers are satisfied with their increasingly automated and mobile are sometimes less familiar with the threats presented by the use of speech interactions with your company. This presentation will talk about the shift technologies. In this session you will learn how to identify emerging threats of power from company to consumer. It’s also designed to show the impact to the speech channel and develop strategies to effectively protect your social media is having on customer care and the opportunity businesses data, customers, and organization. have to make favorable impressions by leveraging the power of speech. Improving Security for Telephone-Based The Critical Importance of Caller Experience Analysis Self-Service Interactions Michael Codini, Chief Technical Officer and Co-Founder, VoiceObjects Steven Cawn, Sales Leader, IBM The business success of automated voice services depends significantly This session will cover current capabilities and market trends related to on a well-founded understanding of how these services are used by cus- improving security for self-service environments. We will discuss such top- tomers. Modern tools for dialogue analysis allow business units to produce ics as federal regulations that are driving deployments (such as multifactor meaningful reports that provide a 360-degree view of customer interac- authentication), current technology capabilities, and industry trends. Case tions. These reports help call centers control and optimize applications and studies will highlight uses of security technologies and considerations for systems and are an essential part of achieving broad corporate objectives, production deployments. including greater customer satisfaction, cost reductions, and increased rev- VoIP Phishing: The Next Threat to IVRs Kris Herrin, Chief Security Officer, Heartland Payment Services Passport for Prizes Identity theft costs us an estimate $53 billion annually, and VoIP phishing is simply the latest assault by hackers to steal personal information. It exploits the trust people have in IVRs, toll-free numbers, and strong brands. Don’t forget to have your “passport” stamped at participating In this presentation, we will witness a live demonstration of VoIP phishing, exhibitors for a chance to win prizes! get under the hood to see exactly how hackers accomplish it, and explain Questions? See any SpeechTEK representative at the Registration Desk, 4th Floor. what companies should do to reduce the risk to their customers and, ulti- mately, their bottom lines.

26 AUGUST 18–20, 2008 w NEW YORK MARRIOTT MARQUIS, NEW YORK CITY WEDNESDAY, AUGUST 20

ical climate, immigration controversies, and the quality of Spanish applica- A304 The Crystal Ball tions, affect companies’ willingness to offer Spanish, as well as callers’ will- 2:45 p.m. – 3:45 p.m. ingness to use it. This session presents data from existing applications that MODERATOR: Ron Owens, Director, Multimedia Applications PSO, Nortel illustrate the need for improvement. W

PANELISTS: E Daniel Hong, Lead Analyst, Datamonitor Beyond Spanish in the U.S.: Acculturated Hispanics and the D

Judith Markowitz, President, J. Markowitz, Consultants Use of Spanglish N

Bill Meisel, President, TMA Associates Jose Elizondo, Professional Services Manager, Multilingual VUI Design, E Nuance Communications S

Experts in this panel will gaze into the crystal ball and give us their predic- D Many companies offer customers the choice of using Spanish but do not

tions for new developments in speech technologies. Join us for a spirited A

discussion of upcoming trends in speech recognition, speaker biometrics, understand the issue of “Spanglish.” Spanish speakers in the U.S. often Y , and speech analytics. This session will prepare you for what’s next in the mix English and Spanish words in the same sentence, or mix languages to A

world of speech and give you hints on how to be one step ahead of the create new words. This has deep implications for the design of grammars U and user dictionaries, as well as the use of parallel grammars. This session curve in making the most of speech technologies. G

will propose methods for understanding and addressing these issues to U

create a successful Spanish system. S

TRACK B w Empire Complex (7th Floor) T

VUI DESIGN PRINCIPLES & TECHNIQUES LUNCH BREAK 2 12:30 p.m. – 1:45 p.m. 0 Making VUIs Even Better B301 10:45 a.m. – 11:30 a.m. Happy Together? Speech and Touchtone B303 MODERATOR: Fran McTernan, Speech Operations Lead, Nortel 1:45 p.m. – 2:30 p.m. This session brings together two different but very valuable topics in VUI MODERATOR: Matt Whipple, Principal Consultant, Self-Service Solutions, design. In the first presentation, we explore yes/no questions—an extremely Avaya common but troublesome element in VUI design. The second presentation Can one user interface encompass both speech and touchtone input in an changes gears to discuss the ways in which VUI design must change to meet efficient, effective, and satisfying interaction? This session explores this the needs of internal-facing VUIs, meant for users within an organization, ver- question and offers insight into the ways speech and touchtone can work sus more common, external-facing, public VUIs. These presentations ask spe- in concert. Learn techniques for writing prompts and call flows that make cific questions and give targeted answers that you can apply in your VUI. sense across modalities, and understand that data supports these tech- niques. This session will teach you how to use both speech and touchtone A Thousand Ways to Yes: Robust Dialogue Design to their best advantage, allowing your organization to meet its business to Recognize Responses to Yes/No Questions objectives. Patrick Nguyen, Chief Technology Officer, Voxify A large variety of yes/no questions are fundamental in today’s speech appli- The Best of Both Worlds: Speech and Touchtone in Harmony cations. Callers offer widely differing responses, which can be a challenge David Attwater, Senior Scientist, Enterprise Integration Group to commonly used, generic yes/no grammars. In this session you will learn This session presents research on user interfaces that combine the stabil- about research into the types of caller responses, implications for dialogue ity of touchtone with the elegance of speech recognition. With results from design, and how to build robust templates that can handle the variation. real-world testing, you’ll learn how modality preference varies among callers, by context for an individual caller, and even on a turn-by-turn basis in the Does One Size Fit All? dialogue. We present a behaviorist approach to show how interfaces can Bob Cooper, Chief Architect, Voice Portal, Avaya rapidly adapt and with no overhead in terms of task completion time. Are there a different set of rules that should be followed for a private-facing application versus a public-facing one? This discussion will focus on les- What’s Up With “Press or Say”? sons learned in designing a personalized, private-facing application, detail- Phil Shinn, Principal Consultant, Voice Application Services, Genesys ing where its design overlaps with a public-facing app and where the needs Telecommunication Laboratories of private users are clearly different. You will also hear how presence and This presentation reviews the pros and cons of designing and deploying a multimodality may affect where private-facing applications could be headed. “press or say” speech application. While some believe this is an easy way to transition from DTMF to speech applications in the enterprise, others Designing for Spanish B302 consider it a bad idea for a variety of reasons. This session will detail recent 11:45 a.m. – 12:30 p.m. findings to help you make the best choice. MODERATOR: Melanie Polkosky, Human Factors Psychologist and Consultant, IBM The Ultimate VUI Showdown B304 In too many U.S.-based speech projects, the Spanish-language portion of 2:45 p.m. – 3:45 p.m. the project is an afterthought. Organizations recognize the need to offer MODERATOR: Gregory Simsar, Vice President, Speech Services, telephony services in Spanish for their growing Hispanic customer base, Syntellect but do not always devote the resources and attention necessary to ensure PANELISTS: a positive and productive experience for these customers. The experts in Bruce Balentine, EVP and Chief Scientist, Enterprise Integration Group this session will present proactive strategies that ensure your Spanish voice Ahmed Bouzid, Senior Product Manager, Angel.com interaction meets the unique needs of Hispanic customers, offering them Wally Brill, SVP of Creative Services, VoxGen Group a compelling reason to use automation. Lizanne Kaiser, Customer Experience Designer, Genesys Telecommunication Laboratories Para Continuar en Espanol: Best Practices for Offering Spanish Peter Krogh, Speech Mechanic, SpeechCycle in U.S. Telephony Applications Elizabeth Strand, Director of User Experience Design, Tellme, a subsidiary Sondra Ahlen, Principal VUI Consultant/Owner, SAVIC of Microsoft A caller’s initial interaction with a multilingual application is in choosing the What are the hot topics in VUI today? Where do we agree and where are language for communicating with the system, typically English or Spanish. there still controversies? Join us for the third annual panel debate on the The presentation of language choice can affect system performance, cus- state of the art in VUI design, refereed again this year by Greg Simsar. Come tomer satisfaction, and company image. Diverse issues, such as the polit- with questions for this year’s panel of VUI practitioners and be ready to

WWW.SPEECHTEK.COM 27 share your 2 cents in spirited discussions of where VUI stands as a profes- on their personal computers. This talk will show how speech-enabled self- sion and discipline. service applications can become even more effective by displaying to a caller a visual display or video at the same time that they use the speech

0 TRACK C Majestic Complex (6th Floor) application menus, pictures of products, live video clips, and video com- 2 w TODAY’S SPEECH TECHNOLOGIES mercials. Specific application development techniques that leverage speech T and visuals together will be presented, along with sample speech-enabled S C301 U Video and Speech voice-plus-video applications for universities, travel, retail, and home health-

G 10:45 a.m. – 11:30 a.m. care. The process of application authoring will be part of the talk.

U MODERATOR: Matthew Yuschik, Human Factors Specialist, Multichannel

A C302

Self Care, Convergys Speech in Automotive ,

Y Around the world, consumers have access to live, two-way video using 3G cell 11:45 a.m. – 12:30 p.m. A phones, broadband/DSL videophones, and video telephone software on their MODERATOR: Paolo Baggia, Director of International Standards, D computers. Speech-enabled self-service applications can become even more Loquendo S

E effective by displaying a visual display or video at the same time they use the Speech technology is increasingly being used in the automobile. The first

N speech application. The phone display may present menus, pictures of prod- case study in this session will show how a speech-enabled, in-vehicle device ucts, live video clips, and video commercials. This session discusses how to D can read email and text messages aloud and enable users to instantly

E harness video in VXML to quickly create rich voice and video applications, as respond by voice. This session’s second case study will reveal how cus- well as how to author multimodal applications for these devices. W tomers can use their mobile phones to retrieve location and current real- Voice, Video and VXML: Add Visual Va-Va-Voom to Your VUI! time parking availability at parking facilities; make parking reservations at Andrew Fuller, CTO, Voxpilot a facility; pay for parking; and provision a new user account and provide With the increasing proliferation of fixed and mobile videophones and outbound informational alerts. media-friendly next-generation networks such as IMS coming closer to a reality, the market waits for 1-D audio services to blossom into interactive, Success Story: iLANE, the Voice-Enabled Travel Companion personable, engaging multimedia interfaces. In this presentation we show CCS Ben Miners, Product Manager, Intelligent Mechatronic Systems how to effectively harness video-in-VXML to quickly create rich voice and A new, state-of-the-art, speech-enabled in-vehicle device exists that not video applications and discuss some considerations when implementing only reads email and text messages aloud, but enables you to instantly a wide range of new services such as video mail, adult services, video dat- respond to them by voice. Generated speech notifies drivers with timely ing, TV-on-mobile, video contact centers, etc. information about phone calls, calendar appointments, and other poten- Techniques for Developing Speech tially distracting mobile phone events. The following challenges will be Plus Video Self-Service Applications covered: In-vehicle voice interfaces require extremely careful design and focus around the user’s primary task—driving; and flexibility, personaliza- Valentine00_iFLYTEK_HalfH Matula, Director Multimedia 7/15/08 R esear 3:04ch, AvayaPM PageLabs 1 Around the world consumers have access to live two-way video using 3G tion, and appropriateness of dialogue content are important in a VUI used cell phones, broadband/DSL videophones, and video telephone software on a daily basis.

28 AUGUST 18–20, 2008 w NEW YORK MARRIOTT MARQUIS, NEW YORK CITY WEDNESDAY, AUGUST 20

Where Does Speech Fit Into Location-Based Services? Multimodal Dialogue Strategies CCS Christian McCarrick, CTO, ParkingCarma, Inc. Dave Pelland, Director, Design Collaborative, Intervoice ParkingCarma needed a scalable and safe way to enable its customers to There are several different strategies for designing multimodal dialogues W access its services while driving. Companies should be able to perform the for use on the handset. Some appear to be less multimodal and more mul- following functions in as cost-effective and convenient manner possible timedia while others achieve true multimodal interaction. This talk will pres- E through a speech enabled application over a mobile phone: search for and ent different strategies along with the pros and cons for each and how to D get the location and current real-time parking availability at parking facili- determine which strategy might be the right one for your application. N ties, make a reservation for parking at a facility, pay for parking at a facility, E Conducting and Evaluating a Multimodal Usability Test S and provision a new user account and provide outbound informational D Cliff McEnery, UI Designer, Nuance Communications alerts. The system also had to be designed to take into account the harsh A noise environment inside of cars and parking facilities. As multimodal applications become more widely available, planning for and Y ,

conducting usability studies are critical steps for improving the ease-of-use of these applications. This presentation will discuss strategies for testing A LUNCH BREAK U the usability of multimodal applications, including how to set up the test 12:30 p.m. – 1:45 p.m. G environment and some of the special usability concerns a multimodal appli- U cation presents. Lessons learned from an on-site multimodal usability study Multimodal Standards and Applications C303 S

will be included as well. T 1:45 p.m. – 2:30 p.m.

2

MODERATOR: Andrew Fuller, CTO, Voxpilot 0 This presentation will showcase standards-based multimodal applications TRACK D w Shubert Complex (6th Floor) developed by member companies of the W3C Multimodal Interaction Work- DEVELOPMENT AND DEPLOYMENT ing Group. It will also demonstrate some laboratory applications that illus- trate the principles of the W3C Multimodal Architecture and a deployed Speech in Financial Services/Insurance D301 healthcare application that enables users to type, speak, and scribble on 10:45 a.m. – 11:30 a.m. their mobile devices to record vital readings, highlight regions of interest on MODERATOR: Rob Marchand, Senior Director, Product Management, images, and submit them wirelessly. Genesys Telecommunication Laboratories Multimodal Applications Using W3C Standards Financial services and insurance companies are turning to speech technol- Deborah Dahl, Principal, Conversational Technologies ogy for help in meeting stringent security, privacy, and compliance Ingmar Kliche, Project Manager, T-Systems Enterprise Services GmbH demands. The first case study will show how a financial services company This presentation will showcase standards-based multimodal applications uses a real-time speech solution that recognizes callers who typically call that illustrate the principles of the W3C Multimodal Architecture and how for specific reasons and proactively engages in data-driven dialogues to the emerging multimodal standards being developed by the group can be offer information they usually request. The second case study will highlight applied to integrate speech and GUI technologies and simplify the creation how an insurance company provides members with prompt access to infor- of interoperable multimodal applications. mation using speech technology, which helped to cut costs, reduce call vol- umes, and maintain a high level of customer satisfaction. Mobile Multimodal Patient Monitoring Raj Tumuluri, President, Openstream State-of-the Art Call Center Openstream has built a suite of healthcare applications incorporating its CCS Brent Sparks, Vice President, Self-Service Channels, Wachovia Bank multimodal SmartCare technology for the use of care-givers, physicians, Wachovia uses a real-time, rules-based, customer-segment-driven decision- and patients who are self-monitoring. This healthcare application combines ing solution that recognizes callers who typically call for specific reasons speech and visual images, allowing users to type, speak, and scribble on and proactively engages in data-driven dialogues to offer information they their mobile devices to record vital readings, highlight regions of interest on usually request. This approach gives the customer information faster and images, and submit them wirelessly. The system stores the images, over- bolsters IVR containment while simultaneously improving the customer lay —gestures, and spoken comments — and plays them upon retrieval, experience. This session will recap the approach, solution, and current allowing other experts to review/add their comments and recommenda- results, allowing attendees to learn more about this data-driven decisioning tions in a similar way. technology and how the IVR can come full circle in providing personal treat- ment. Developing and Testing Multimodal Applications C304 Leveraging Speech Technology in Healthcare 2:45 p.m. – 3:45 p.m. CC Remus Siclovan, Senior Systems Analyst, Health Net MODERATOR: Hamilton Yu, Manager of Client and Product Delivery, S Tellme, a division of Microsoft Health Net realized cost savings between $2.3 million and $4.4 million annu- ally using speech technology to automate processes, reduce call volumes, There are several different strategies for designing multimodal dialogues for and maintain a high level of customer satisfaction. This session will discuss use on handsets. This session will discuss the pros and cons for each strat- how speech technology can help healthcare providers balance cost effec- egy and how to determine the best strategy for your application. Other topics tiveness and service goals while providing members with prompt access of discussion include how to conduct usability studies, a critical step in improv- to information and how to enhance security, privacy, and HIPAA compli- ing an application’s ease of use, how to set up a test environment, and some ance using voiceprint identification. of the special usability concerns a multimodal application presents. Testing D302 Presentation Links 11:45 a.m. – 12:30 p.m. MODERATOR: Emmett Coin, Speech Scientist, ejTalk This session addresses two important testing problems. The first is capac- Get access to many PowerPoint presentations and ity testing. Most businesses have peak periods of high activity throughout websites used in conjunction with the SpeechTEK the year. Learn how to define and implement testing that will help mitigate the risk of implementing a system that is not ready for peak-period deploy- 2008 sessions! Links will be posted on the ment. The second is using open source software for testing. This session SpeechTEK.com website soon after the conference. will review an integrated, open standards application tuning-based approach based on Eclipse. Application reporting, performance application logs, tun-

WWW.SPEECHTEK.COM 29 WEDNESDAY, AUGUST 20

ing components, as well as how the latest approaches to integrated tuning that indentify call types, patterns, and repeat customers drives customer tools complement the application tuning process will also be covered. calls to quick resolutions. This session will explore using speech applica- tion, customer surveys, and caller analysis to understand the voice of the Performance Testing for Peak Periods

0 customer and meeting customer needs in a high-speed environment.

2 Jim Jenkins, CEO, IQ Services Self-Service Using Telephony-Based Order Taking

T Most businesses have peak periods of high activity throughout the year.

S Making sure that your contact center solution infrastructure and related CCS Richard Grant, CTO, OrderCatcher LLC U technologies will work as required under these peak periods is a challenge. OrderCatcher is a cost-effective self-service speech application for Chinese G In this session the attendees will learn how to define and implement test- and Pizza take-out. Learn why customers chose speech self-service, the U ing that will help mitigate the risk of implementing a system that is not ready business problem they wanted speech self-service to solve, how they devel- A

for peak period deployment. oped their requirements for the speech self-service application, and what ,

Y advice they would give to a company considering speech self-service to

A Consolidate and Correlate: take their fast-food orders over the phone.

D Integrated Application Design and Tuning

S Judi Halperin, Speech Engineer, Contact Center Practice, Self Service

E D304 Solutions, Avaya Tuning Tips N Dennis Jakobsen, MTS, Avaya 2:45 p.m. – 3:45 p.m. D Tuning is an absolutely critical step in the success of any speech imple- MODERATOR: Emmett Coin, Speech Scientist, ejTalk E mentation. Today speech application developers use various application Tuning a speech recognition application results in a more accurate solu- W development packages but often rely on separate and costly commercial tion that can improve the caller experience. The process involves using tools or tool scripts for application tuning. This session will review an inte- prompt, grammar, call flow, and caller data to help developers make the grated open standards-based approach based on Eclipse, which simpli- needed changes. While nearly 40%–50% of total development and deploy- fies direct correlation between the design of the application and its real- ment time is often spent on application tuning, there are techniques to cre- world performance. Application reporting, performance application logs, ate a repeatable process that reduces time and gives the desired results. tuning components, as well as how the latest approaches to integrated tun- Two experts will present their favorite tips and techniques for effectively tun- ing tools complement the application tuning process will be covered. ing speech applications.

LUNCH BREAK Tuning Speech Applications Yves Normandin, President, Nü Echo 12:30 p.m. – 1:45 p.m. With the help of numerous real-world examples, this session will present a D303 systematic approach to the task of tuning speech applications using field- Speech in Customer Service collected utterances. Areas covered by the tutorial include effective gram- 1:45 p.m. – 2:30 p.m. mar development and maintenance, the process of tuning grammars using MODERATOR: Paolo Baggia, Director of International Standards, field utterances, key metrics to measure grammar performance, techniques Loquendo to efficiently manage out-of-grammar utterances, and tuning confidence Speech applications are a driving force in defining the customer care expe- scores and setting confidence thresholds. rience. Designing and deploying automated speech solutions from the cus- tomer’s point of view is a critical factor for competitive advantage. Learn Speech Tuning, Practical Tips to Improve Performance the value from using speech to optimize your customer interactions to drive Stephen Keller, Project Manager, LumenVox higher business performance, increase automation rates, and improve your While nearly 40%–50% of total development and deployment time is often customer satisfaction. This session will show the value of understanding spent on application tuning, there are techniques to create a repeatable your speech applications and delivering a rewarding customer experience. process that reduces time and gives the desired results. This presentation will address how to effectively tune speech applications by presenting prac- Delivering High-Speed Customer Service tical tips and proven techniques. Chris Weeks, Division Vice President, Customer Care, Comcast Delivering customer service in the most effective and efficient way is nec- essary for a successful customer service platform. Developing applications

8:00 a.m. – 8:40 a.m. Wednesday Sunrise Semester Discussions w SUNRISE SEMESTER DISCUSSION 8 Royale Room (6th floor) Analytics DISCUSSION LEADER: SUNRISE SEMESTER DISCUSSION 6 SUNRISE SEMESTER DISCUSSION 7 Aphrodite Brinsmead, Associate Analyst, Datamonitor Speech Development Teams Project & Perfect: Using Success Datamonitor developed a method to help DISCUSSION LEADER: Rate Projection Techniques to Make determine where to focus further work by Nava Shaked, CRM and CC Leader, IBM Better Tuning Decisions projecting the likely impact of changes on Join us for an overview of the differences between DISCUSSION LEADERS: success rates in a format that is clear to a traditional software development team and a Louise Tranter, Lead Speech Applications & understand for our customers. The method speech application development team. Possible Sarah Turney, Senior Speech Applications also allows us to highlight issues with the dif- discussion topics will include the following: Specialist, Fluency Voice Technology ferent aspects of the application in its live • How to staff a speech application development A customer concerned that their application was environment, including recognition rates, team (e.g., hire consultants temporarily, hire new not performing drove us to explore new ways of data quality and access and customer employees, or retrain existing employees) analyzing system performance and conveying the behavior. We have been able to use this with • How to improve communication among team results to our customers. the customer to ensure that the focus of members efforts on both sides is targeted at where it • What to do when projects are behind schedule can be most productive.

30 AUGUST 18–20, 2008 w NEW YORK MARRIOTT MARQUIS, NEW YORK CITY AD SPEECHTEK LABS

This year marks our first SpeechTEK Labs, in which attendees can sample the latest speech technology products, talk to developers, and speak openly to our expert panel of judges about the technology. These sessions, which take place in the Exhibit Hall on Tuesday and Wednesday, will provide objective evaluations of new speech technology and show how speech is breaking new ground in various indus- tries. Attendees will have the opportunity to evaluate speech products in three categories: multimodal applications, speech tools, and voice- activated games. As an added benefit, after the lab, the judges will issue ratings of the tools to provide further guidance.

TUESDAY, AUGUST 19 munications solutions featuring voice, video, web, fax and email that easily integrate with a wide range of databases, legacy, and back-office systems. 10:45 a.m. – 12:30 p.m. Loquendo TTS Director: The Next Generation Suite of TOOLS FOR SPEECH APPLICATIONS Authoring Tools for Speech Synthesis MODERATOR: Moshe Yudkowsky, President, Disaggregate Consulting It is a complete authoring tool for creating, testing and refining voice What tools are available for speech technology application developers? prompts. It will simplify the creation of effective prompts for any of your appli- Developers at all skill levels can now select from a variety of competing and cations in a few simple steps, but it also facilitates the use of our innovative complementary tools for the entire development process, from initial test- features, such as emotional TTS, the reading style definition, as well as the ing through coding through final testing. Visit this lab session to see what’s new voices creation and the user-driven unit selection tools. available and to speak one-on-one with the tool developers. LumenVox Speech Tuner Ajax Weaver The LumenVox Speech Tuner is designed to performs tuning and tran- VoiceXML Orchestrator is the very first web-based drag-and-drop environ- scription, as well as instant parameter, grammar, and version upgrade test- ment that allows end users to build self-service IVR applications inside Fire- ing of any speech recognition application. It reduces the work of your post- fox and IE. It offers a sleek Ajax Weaver interface and WYSIWYG capabili- deployment application revisions and allows you to bring tuning in-house, ties to visually layout call flows in a manner similar to creating Visio diagrams. avoiding costly professional service fees. No software install, no downloads, no plugins. VocaLabs Avaya Dialog Designer VocaLabs will be demonstrating its intuitive real-time usability survey reports, The Avaya Dialog Designer is an open standards-based Integrated Devel- which allow VUI designers and clients to quickly discover ways to improve opment Environment (IDE) for Avaya Voice Portal and Avaya Interactive application designs and increase user acceptance. Response. Dialog Designer accelerates time to market for businesses seek- ing higher automation and closure rates, turns a faster return on investment Voiyager Come and see firsthand how Voiyager is revolutionizing the VoiceXML devel- for Avaya Self Services solutions, and delivers superior customer satisfac- opment lifecycle. Learn how Voiyager can automatically explore, test, and tion at a lower investment level. Based on the widely accepted Eclipse.org validate your service applications in a more efficient and time-saving man- development framework for fit in information technology (IT) departments, ner. No other testing methodologies or products provide enterprises with Dialog Designer offers web application developers a reusable drag-and- the validation that deployed applications are error-free. drop environment for development and maintenance of speech and touch- tone IVR applications. Voxeo Cisco Unified Call Studio 7.0 West Interactive Cisco Unified Call Studio 7.0 is an integrated development environment (IDE) for voice self-service applications that allows you to build, debug, man- age, and document sophisticated speech-enabled telephone self-service 1:45 p.m. – 3:30 p.m. applications for Cisco and non-Cisco VoiceXML platforms. VOICE-ACTIVATED GAMES Envox Communications Development Platform MODERATOR: David Thomson, CTO, SpeechPhone This is an advanced graphical development environment that reduces IVR See some of the newest titles in voice-activated video games, such as Tom development and deployment time by 50% or more compared to traditional Clancy’s End War. Visitors will get a chance to try the games and get a first- graphical development environments. Standardize on the Envox Communi- hand look at how speech technology lets players give verbal commands to cations Development Platform to quickly create multichannel IVR and com- virtual teammates that go beyond the limitations of the game controller.

WEDNESDAY, AUGUST 20 10:45 a.m. – 12:30 p.m. Intervoice Voice Portal and Wireless Carriers’ 3G Networks After receiving a flight cancellation SMS or email alert, an airline passen- MULTIMODAL APPLICATION LAB ger may use his/her voice and touch screen interchangeably to reserve a MODERATOR: Deborah Dahl, Principal, Conversational Technologies flight, including pricing options and seat selection. If the passenger requires See how voice can be combined with other input modalities to dramatically more information, he/she can touch a button or say “Agent” to transfer to enhance the capabilities and usability of applications. This is your chance a Contact Center Agent. to see the future of speech applications and learn about the effectiveness Loquendo: Voice Multimodal Application Framework— of real multimodal applications. The Next Generation for Multimodal Voice Interaction Avaya VoiceVerified IQ Services Integrated Multichannel Testing Our demo center is a CTI integrated experience so participants are on the Contact center and unified communications solutions allow customers to internet and the phone at the same time and can see the screen change make contact via different devices. Is your contact center infrastructure ready when they are authenticated. The IVR stand-alone demo is a very “hands to support these diverse communications? Integrated Multichannel Testing on” demo that is sure to get a “wow” from the participants. from IQ Services gives you confidence the critical elements of your solution work separately and together to deliver a better customer experience.

32 AUGUST 18–20, 2008 w NEW YORK MARRIOTT MARQUIS, NEW YORK CITY AD Sondra Ahlen Mike Bergelson Aphrodite Brinsmead Principal VUI Consultant/Owner Director, Product Management Associate Analyst SAVIC Cisco Datamonitor 209-962-6444 646-674-7450 [email protected] Y [email protected] [email protected] R Jenny Burr O Joe Alwan Jonathan Bloom Manager, Speech Technology, T Vice President & General Manager Senior Interaction Designer Global Consulting Services C

E BBN Technologies SpeechCycle Intervoice R

I 617-873-7634 646-792-2213 972-454-8794

D [email protected] [email protected] [email protected]

R

E Lior Arussy Ingo Bors Elaine Cascio

K President Industry Consultant Vice President

A Strativity Group Ibolstar Consulting Services Vanguard Communications Corp. E

P 973-796-5310 650-288-0299 401-849-9099

S [email protected] [email protected] [email protected]

David Attwater Greg Borton Steven Cawn Senior Scientist VP of Analytics Sales Leader Enterprise Integration Group Nuance Communications IBM +44 797 1244038 415-693-8646 877-245-2127 [email protected] [email protected] [email protected]

Alexandra Auckland Dominique Boucher Steve Chirokas Sr. VUI Designer Lead Software Developer VP, Marketing Nuance Communications Nü Echo Inc. SpeechCycle 212-785-2272 514-861-3246 917-575-5371 [email protected] [email protected] [email protected]

R.J. Auburn Ahmed Bouzid David Claiborn Chief Technology Officer Senior Product Manager Senior Designer Voxeo Angel.com United Healthcare 407-418-1800 703-714-1078 913-226-1788 [email protected] [email protected] [email protected]

Paolo Baggia Susan Boyce Michael Codini Director of International Standards Principal Designer Chief Technical Officer & Co-Founder Loquendo Tellme Networks, Inc. VoiceObjects, Inc. +39 011 291 3425 732-219-7646 650-288-0299 [email protected] [email protected] [email protected]

Bruce Balentine Monique Bozeman Jordan Cohen EVP & Chief Scientist Marketing Consultant Senior Scientist Enterprise Integration Group Monique Bozeman Consulting SRI 940-891-3414 510-393-1932 339-298-1008 [email protected] [email protected] [email protected]

James Barnett Wally Brill Emmett Coin Senior Technical Director SVP of Creative Services Speech Scientist Aspect Software VoxGen Group ejTalk 978-250-7528 [email protected] 724-940-7000 [email protected] [email protected]

Bob Cooper Evening Receptions Chief Architect Voice Portal Avaya 803-231-2714 SpeechTEK invites attendees with a Networking Pass or above to a Welcome [email protected]

Reception held Monday evening in the Exhibit Hall from 5:00 p.m. – 7:00 p.m. Bonnie Crater sponsored by Angel.com as well as the Tuesday evening reception from SVP Marketing 5:30 p.m. – 7:00 p.m. on the 9th floor. Mingle with exhibitors, speakers, and VoiceObjects, Inc. 415-265-1804 conference attendees while enjoying a glass of wine and light hors d'oeuvres. [email protected]

34 AUGUST 18–20, 2008 NEW YORK MARRIOTT MARQUIS, NEW YORK CITY SPEAKER DIRECTORY

Deborah Dahl Leah Eyler Charles Giordano Principal Product Manager Associate Director Conversational Technologies SpeechCycle Bell Canada

610-888-4532 646-576-6691 905-282-3164 S [email protected] [email protected] [email protected] P E A

Ricardo de Carvalho Destro Nick Ezzo Glen Graham K

Technology Director Director of Marketing SVP Business Operations E Volans Technology Tu Vo x Bank of America R

PH: +55 11 92346650 408-625-1617 704-683-2637 D I [email protected] [email protected] [email protected] R E

Rita Dhruve Andrew Fuller Richard Grant C Speech Solutions Specialist Chief Technology Officer Chief Technology Officer T O Nuance Communications Voxpilot OrderCatcher Inc. R

732-766-9385 +353-1-209-1919 305-444-8226 Y [email protected] [email protected] [email protected]

Jose Elizondo Charles Galles Hannah Grap Professional Services Manager, Multimedia Applications Speech Analyst Marketing Communications Manager Multilingual VUI Design Nortel Networks Language Weaver, Inc. Nuance Communications [email protected] 310-437-7300 617-557-4401 [email protected] [email protected] Marsal Gavalda Director of Incubation Judi Halperin David Englehardt Nexidia Speech Engineer President 404-495-7220 Avaya Advanced Technical Support [email protected] 407-838-4181 [email protected] [email protected] Juan Gilbert T-SYS Distinguished Associate Professor 00_Angelcom_HalfH_STEK_08 7/16/08Auburn 11:46 University AM Page 1 334-844-6316 [email protected]

Angel.com powers smarter voice applications for businesses large and small. Our powerful IVR and call center technologies help make businesses more ecient, Smarter voice applications and your callers get the information they need faster. Seamlessly integrate service and support lines with CRM Welcome back, Jane, would you like to update your systems and other backend databases to better serve your customer support ticket? customers. And, Angel.com solutions are built on a web-based platform making it quick and easy to implement a voice application that meets your unique business needs, while at the same time reducing costs.

What’s smarter than that? Want to nd out how Angel.com can make your business smarter? Call us at 888-692-8165 or visit us online at www.angel.com.

WWW.SPEECHTEK.COM 35 Kris Herrin Manohar Kesireddy Judith Markowitz Chief Security Officer Business Solutions Architect President Heartland Payment Systems, Inc. Verizon J. Markowitz, Consultants 972-454-8000 972-507-1657 773-769-9243

Y [email protected] [email protected] [email protected] R

O Daniel Hong Ingmar Kliche David Martin T Lead Analyst Project Manager Principal Consultant-Speech Engineering C Datamonitor E T-Systems Enterprise Services GmbH Avaya

R 212-652-5375

I +49 30 3497 2348 321-663-1950 [email protected]

D [email protected] [email protected]

R Phillip Hunter

E Peter Krogh Valentine Matula

K Vice President, Interaction Design Speech Mechanic Director, Multimedia Research

A SpeechCycle SpeechCycle Avaya E 469-853-9016 530-265-2392 614-825-6415 P

S [email protected] [email protected] [email protected]

Susan Hura Ray Kurzweil Christian McCarrick Principal Author and Keynote presenter CTO SpeechUsability Kurzweil Technologies Inc. ParkingCarma, Inc. 404-702-4723 781-263-0000 650-281-9864 [email protected] [email protected] [email protected]

Sunil Issar Cliff LaCoursiere Cliff McEnery Director SVP Business Development UI Designer Convergys CallMiner Nuance Communications 703-885-5753 617-899-1433 415-371-8255 Ext 222 [email protected] [email protected] [email protected]

Michael Jacks James Larson Jenni McKienzie Senior Manager, Logistics & Vice President Senior Business Solutions Advisor Transportation Larson Technical Services Travelocity Coca-Cola Enterprises 503-645-3598 682-605-4069 770-370-7368 [email protected] [email protected] [email protected] Ethan Levine Fran McTernan Dennis Jakobsen Voice Interaction Designer Speech Operations Lead MTS SpeechCycle Nortel Networks Avaya 407-574-7282 732-796-2025 408-577-7167 [email protected] [email protected] [email protected] Charles Lewis Bill Meisel Jim Jenkins Senior Technical Trainer President CEO Bloomberg LP TMA Associates IQ Services 212-260-4084 818-708-0962 612-243-5123 [email protected] [email protected] [email protected] Jackson Liscombe James Mesbur Lizanne Kaiser Speech Engineer Voice Interaction Designer/ Customer Experience Designer SpeechCycle Speech Scientist Engineer Genesys Telecommunications [email protected] SpeechCycle Laboratories 646-792-2183 510-965-0528 Fred MacKenzie [email protected] [email protected] Senior Business Solutions Advisor Bell Canada Dan Miller Rob Kassel 416-353-0826 Senior Analyst Vice President, Marketing & Product [email protected] Opus Research, Inc. Holly Connects 415-904-7666 857-241-3717 Rob Marchand [email protected] [email protected] Senior Director, Product Management Genesys Telecommunications Laboratories Jim Milroy Stephen Keller 905-968-3304 Director, Creative Services Project Manager [email protected] West Interactive LumenVox 402-965-7422 858-707-7700 [email protected] [email protected]

36 AUGUST 18–20, 2008 NEW YORK MARRIOTT MARQUIS, NEW YORK CITY SPEAKER DIRECTORY

Ben Miners Daniel Padgett Whitney Quesenbery Product Manager Senior VUI Consultant Consultant/Researcher Intelligent Mechatronic Systems VoicePartners WQ Usability

519-745-8887 773-942-6946 908-638-5467 S [email protected] [email protected] P E

Mary Constance Parks A

Karen Moylé Senior Voice User Interface Designer Ken Rehor K

Voice Interaction Designer Nuance Communications Founder E SpeechCycle 415-260-1098 VoiceXML Forum R

646-826-2327 [email protected] 408-275-6488 D I [email protected] [email protected] R

Tim Pearce E Tim Moynihan Global Solution Manager, Self Service Michael Roberts C Vice President, Global Marketing & Dimension Data Business Integration Manager - Telecom T O Sales Support +44 7909 976 030 NSTAR R

Envox Worldwide [email protected] 781-441-3377 Y 508-871-7656 [email protected] [email protected] Dave Pelland Director, Design Collaborative Renae Rogers Carrie Nelson Intervoice Dialogue Designer Speech Solutions Team Technical Lead 972-454-8000 West Corp. Nortel Networks [email protected] 800-542-1000 631-285-2101 [email protected] [email protected] Jessica Peterson Speech Technology Consultant Thomas Schalk Val Nenov Versay Solutions Vice President, Voice Technology Adjunct Professor 888-869-0121 ATX UCLA Medical Center [email protected] 972-753-6278 310-498-6197 [email protected] [email protected] Mark Pfeiffer Vice President Business Development & Bill Scholz Patrick Nguyen Communications President Chief Technology Officer SAIL LABS Technology AG NewSpeech LLC Voxify +43 1 58095-580 610-212-8016 510-545-5000 [email protected] [email protected] [email protected] Michael Picheny Tom Scott Yves Normandin Manager CIO & Sr. VP Operations President IBM Spiegel Brands Nü Echo, Inc. 914-945-2527 757-825-4182 514-861-3246 Ext 225 [email protected] [email protected] [email protected] Roberto Pieraccini Vlad Sejnoha John Oh Chief Technology Officer Vice President, Chief Scientist Technical Lead SpeechCycle Nuance Communications Cisco 646-792-2744 781-565-5164 646-674-7421 [email protected] [email protected] [email protected] Melanie Polkosky Nava Shaked Eduardo Olvera Human Factors Psychologist & Consultant CRM & CC Practice Leader Senior User Interface Designer IBM IBM Nuance Communications 901-861-9657 +972 (3) 916 5483 480-206-0966 [email protected] [email protected] [email protected] Steven Pollock Phil Shinn Daniel O’Sullivan Executive VP & Co-Founder Principal Consultant President/CEO Tu Vo x Voice Application Services Interactive Digital 408-625-1617 Genesys Telecommunications Laboratories 631-680-4307 [email protected] 310-488-9267 [email protected] [email protected] Matt Prather Ron Owens Staff Engineer, VUI Design Roberto Sicconi Director Intervoice Program Director Nortel Networks 415-317-4995 IBM 402-222-7019 [email protected] 914-945-3159 [email protected] [email protected]

WWW.SPEECHTEK.COM 37 SPEAKER DIRECTORY

Remus Siclovan Louise Tranter Chris Weeks Senior Systems Analyst Lead Speech Application Specialist Division Vice President, Customer Care Health Net, Inc. Fluency Voice Technology Comcast 818-676-6541 + 44 (0) 20-3178-5618 720-267-7584 Y [email protected] [email protected] [email protected] R

O Gregory Simsar Peter Trompetter Matt Whipple T Vice President, Speech Services Vice President, Global Development Principal Consultant, C

E Syntellect, Inc. GyrusLogic, Inc. Self-Service Solutions R

I 602-789-2902 602-432-1995 Avaya

D [email protected] [email protected] 303-538-7787 [email protected] R

E Lynda Kate Smith Darla Tucker

K Vice President, General Manager Principal, Global Consulting Services James Whitten

A Nuance Communications Intervoice Assistant Section Manager E 408-922-6288 214-417-4560 San Francisco Public Utilities Commission P

S [email protected] [email protected] 415-551-3054 [email protected] Brent Sparks Raj Tumuluri Vice President, Self Service Channels President Simonie Wilson Wachovia Bank Openstream, Inc. Senior Speech Scientist 704-427-7544 732-507-7033 VUI Design & Tuning [email protected] [email protected] Intervoice 972-545-8000 Jessica Stevens Sarah Turney [email protected] Senior Engineer, VUI Design VUI Designer Intervoice lastminute.com Dan York 972-454-8000 sarah.turney@ yahoo.co.uk Director of Emerging [email protected] Communication Technology Julie Underdahl Voxeo Elizabeth Strand Senior VUI Design Engineer 407-418-1800 Director of User Experience Design Intervoice [email protected] Tellme Networks, Inc. 972-454-8788 650.930.9302 [email protected] Hamilton Yu [email protected] Manager of Client & Product Delivery Marge Vizcarra Tellme Networks, Inc. David Suendermann Customer Services Manager [email protected] Speech Science Engineer San Francisco Public Utilities Commission SpeechCycle 415-551-4707 Moshe Yudkowsky 646-792-2173 [email protected] President [email protected] Disaggregate Leslie Carroll Walker 773-764-8727 David Thomson Staff Engineer, VUI Design [email protected] Chief Technology Officer Intervoice SpeechPhone 972-454-8371 Matthew Yuschik 949-655-1677 [email protected] Human Factors Specialist, [email protected] Multi-Channel Self Care Mark Webb Convergys David Thornton IVR Process Engineer, VUI Design 513-784-2096 Instructor Humana IVR User Interface Designs [email protected] Auburn University 502-476-2317 256-343-2772 [email protected] Catherine Zhu [email protected] Senior Voice User Interface Design Engineer Intervoice Sunrise Semester Discussions 972-454-8510 [email protected]

SpeechTEK will host early morning discussions on Monday, Tuesday, and Imed Zitouni Senior Research Scientist Wednesday in the Royale Room (6th floor) for business professionals, IBM project leaders, designers, and developers to openly talk about a variety of 914-945-1346 topics. This is a great opportunity for attendees to network, discuss [email protected] challenges, and share ideas with peers and speech technology professionals.

38 AUGUST 18–20, 2008 NEW YORK MARRIOTT MARQUIS, NEW YORK CITY AD EXHIBITOR LIST & FLOOR PLAN

Acclaim Telecom Services, Inc...... 506 Gold Systems ...... 410 Prosodie ...... 303 Accuvoice, Inc...... 520 IBM ...... 201 Resolvity, Inc...... 617 Aculab ...... 303 iFLYTEK ...... 523 SER Solutions, Inc...... 322 Agnitio...... 315 Inference Communications...... 522 SpeechCycle ...... 513 AjaxWeaver...... 615 Intervoice ...... 604 Speech Technology Magazine ...... 317 Angel.com ...... 806 Inventive Labs ...... 303 Syntellect, Inc...... 510 Aspect Software ...... 415 IQ Services ...... 408 TuVox ...... 409 Autonomy etalk...... 414 Loquendo ...... 509 VClarity ...... 303 Avaya...... 312 LumenVox...... 303 Verint Witness Cepstral...... 303 Message Technologies Inc...... 508 Actionable Solutions ...... 310 Cisco ...... 610 MicroAutomation ...... 314 Vicorp ...... 306 ClickFox ...... 609 NeoSpeech ...... 311 Vocal Laboratories, Inc...... 507 Convergys ...... 304 Nexidia...... 803/805 VoiceObjects, Inc...... 800 CRM...... 317 NICE Systems ...... 614 VoiceVerified, Inc...... 417 EMPIRIX ...... 313 Nortel Networks ...... 801 Voiyager ...... 514 Envox Worldwide ...... 511 Nuance ...... 501 Voxeo ...... 804 First Data Voice Services...... 411 Nu Echo, Inc...... 413 Voxify ...... 505 Genesys Telecommunications Performance Technology Laboratories ...... 316 Partners, (PTP) ...... 412 Walsh Media, Inc...... 515 GM Voices...... 307 Persay Ltd...... 321 West Interactive ...... 502

RESIGN 20' BOOTH 701 SPEECHTEK LABS

722 TELEPHONES 720 20'

617 615 609 603 801 803 805 807 ENTRANCE TO EXHIBITS 20' 20' 20' 20' 523 20' 20' 20' 614 610 604 20' 800 804 806 521 616 515 513 511 509 507 505 501 S

522 T E

20' 20' A

20' - M T I A 514 510 508 506 O I

520 N

20' 20' L 20' S

502 R Y 417 415 413 411 409 O T B A B V O E

10' L L E 20' 20'

414 412 410 408 406 20' 20' 20' 322 317 315 313 311 307 303

321 20' 20' 18' 316 314 312 310306 304 20' 20' 10' 10' 223 12' 20' 20' 20' 201 901 903 905 907 ENTRANCE TO EXHIBITS 221 217 213 211 209 205 10' 20'

20' 20' 20' 900 904 906 208 204 214 20' E N

O WOMEN MEN H

111 107 105 101 P E L E T

8' 116 114 112 110 108 106 104 102 100 Part of the CRM Showcase

40 AUGUST 18–20, 2008 w NEW YORK MARRIOTT MARQUIS, NEW YORK CITY AD Acclaim Telecom cessfully adapted its robust technolo- small and medium enterprises to com- Cisco Services, Inc. gy to provide the most sophisticated munications-enable their customer corporate authentication product. service, collections and sales & tele- 170 West Tasman Drive 1300 E. Arapaho, Suite 206 Agnitio's technology is currently marketing business processes. San Jose, CA 95134 Y Richardson, TX 75081 deployed in over 20 countries. KIVOX, PH: 408-526-4000 R PH: 972-331-1700 Agnitio's corporate flagship product, www.cisco.com/go/cvp

O FAX: 972-331-1701 can be adapted very easily into exist- Booth 610

T www.acclaimtelecom.com ing voice and technology infrastruc- Cisco is the world leader in IP-based C Booth 506 tures. Unique capabilities include per- contact center solutions for business- E Acclaim Telecom is a full-service forming voice verification in any lan- es of all sizes. The Cisco Unified R I provider of Microsoft OCS 2007 guage and mixing different voice Customer Contact Solutions portfolio

D Speech Server products. We special- channels. includes both time-division multiplex-

R ize in the analysis, design and devel- ing (TDM)- and IP-based voice and opment of Microsoft speech applica- Autonomy etalk video self- and assisted-service solu- O AjaxWeaver tions using current tools and technolo- 2200 Ross Avenue, Suite 3500

T tions for large and midsize contact I gies. From hosting your existing 63 Alameda Way Dallas, TX 75201 centers. Cisco’s self-service solutions, B

I Microsoft speech application to imple- Ottawa, ON K2J 5B3 Canada PH: 214-981-3100 including Cisco Unified Customer PH: 347-834-2664 FAX: 214-981-3500 H menting a new design from the Voice Portal (CVP), offer a comprehen- ground up, Acclaim Telecom is always FAX: 613-249-7320 www.etalk.com X sive set of natural-language voice self-

E the right choice. www.ajaxweaver.com Booth 414 service capabilities to automate Booth 615 Autonomy etalk delivers enterprise incoming customer inquiries and deliv- AjaxWeaver is the pioneer and leader solutions for multichannel interaction er intelligent, personalized self-service in creating web-based drag-and-drop analysis, real-time information assis- over the phone, enabling customers to Ajax IDE for VoiceXML applications. tance, contact center performance, efficiently and enjoyably retrieve the Our flagship product, VXML Orches- and compliance management. By pro- information they need from the contact Accuvoice, Inc. center. 343 Wainwright Drive, Suite 2E trator, is the very first web-based drag- viding integrated tools for recording, and-drop environment that allows end archiving, analyzing, and retrieving crit- Northbrook, IL 60062 users to build self-service dynamic IVR ical interactions, Autonomy etalk helps PH: 847-559-7272 ClickFox applications inside web browsers. The organizations meet regulatory compli- FAX: 847-559-8010 Orchestrator offers sleek Ajax inter- ance, manage enterprise security and 3445 Peachtree Road, Suite 1250 www.accuvoice.com face and WYSIWYG capabilities to risk, enhance customer support, and Atlanta, GA 30326 Booth 520 visually layout call flows in a manner uncover valuable business intelligence PH: 404-351-8020 Accuvoice is a platinum Avaya developer very similar to creating Visio diagrams. for strategic operations. www.clickfox.com with 17 years of self-service and an IBM No software install, No downloads, No Booth 609 and Nuance NLSR. We are a nationwide plugins. Avaya ClickFox is the only solution that government-certified, women’s business delivers true customer experience entity (WBE) and offer business case 211 Mount Airy Road Angel.com analytics behavior, enabling organiza- planning support for cost/revenue, ROI, Basking Ridge, NJ 07920 tions to see how their customers real- and funding scheduling. Visit our booth 1861 International Drive PH: 908-953-6000 ly interact across all touch points, or come to session, D204, “Speech in McClean, VA 22102 www.avaya.com such as IVR, ACD, web, chat, email, Government/Utilities: Developing Call PH: 703-744-3091 Booth 312 CRM, agents, point of sale systems Center Applications,” featuring the city FAX: 703-770-1321 Deliver exceptional value to your busi- and interactive kiosks. ClickFox of San Francisco, to learn more. www.angel.com ness and customers with video and uncovers opportunities for extending Booth 806 voice-enabled self-service solutions. automation, cost-savings and increas- Aculab Angel.com combines powerful busi- Experience the ease of application cre- ing revenue. For more information, visit our website. 197 First Avenue, Suite 130 ness voice technologies with simpli- ation through Avaya Dialog Designer, Needham, MA, 02494 fied implementation and manage- and leverage the resources of the Avaya DevConnect developer program PH: 781 433 6000 ment. Our open, software-as-a-service platform facilitates information-driven, to jump-start your speech-enabled www.aculab.com personalized communications and application efforts. Avaya is a global Booth 303 transactions, improving business effi- leader in communication systems, Aculab offers developers and service ciency and enhancing caller satisfac- applications and services, serving providers a wide range of hardware tion. More than 1,600 customers use more than 1 million customers from and software for integration into high- Angel.com to more efficiently and small businesses up to more than 90% Convergys performance speech-enabled commu- cost-effectively build and manage call of FORTUNE 500 companies. 201 East Fourth Street nications solutions such as voice por- centers and IVR solutions such as Cincinnati, OH 45201 tals and outbound IVR. Scalable phone surveys, payment lines, and PH: 800-344-3000 media processing resources enable more. Cepstral FAX: 513-458-1315 the production of cost-effective enter- 1801 E. Carson Street, Suite 200 www.convergys.com prise solutions for use in both PSTN Pittsburgh, PA 15203 Booth 304 and IP environments. Aspect Software PH: 412-432-0400 Convergys is a global leader in rela- 300 Apollo Drive www.voiceforge.com/demo Chelmsford, MA 1824 tionship management. We provide Booth 303 solutions that drive more value from Agnitio PH: 978-250-7900 Cepstral specializes in text-to-speech the relationships our clients have with Virgilio, 25 FAX: 978-244-7420 and offers the world’s largest catalog their customers and employees. Ciudad de la Imagen www.aspect.com 28223 Pozuelo de Alarcon of English voices. Cepstral TTS Convergys turns these everyday inter- Madrid, Spain Booth 415 engines power embedded, desktop actions into a source of profit and and server markets. NEW!! VoiceForge strategic advantage for our clients. For PH: +44 (0) 7766338089 Aspect Software, Inc. founded the contact center industry and is now the — a hosted (SaaS) voice platform deliv- 25 years, our unique combination of FAX: +34 91 512 24 18 world's largest company focused sole- ering on-demand audio for telephony, domain expertise, operational excel- www.agnitio.es ly on unified communications for the media sites, virtual worlds, consumer lence, and innovative technologies has Booth 315 contact center. Our all-in-one, IT-ready Internet devices, and web applications. delivered process improvement and Agnitio, the worldwide leader in voice solutions help two-thirds of the FOR- “Fun” character voices available at actionable business insight to clients biometrics for public security, has suc- TUNE Global 100 and a variety of www.voiceforge. com/demo. that now span more than 70 countries

42 AUGUST 18–20, 2008 NEW YORK MARRIOTT MARQUIS, NEW YORK CITY EXHIBITOR DIRECTORY and 35 languages. Convergys is a tomer satisfaction. More than 1.25 mil- GM Voices member of the S&P 500 and has been lion ports of Envox’s hosted and prem- voted a Fortune Most Admired ise-based IVR solutions have been 2001 Westside Parkway, Suite 240

Company for 7 consecutive years. We deployed globally and more than 1 mil- Alpharetta, GA 30004 E have approximately 75,000 employees lion contact center agents benefit from PH: 770-752-4500 X in 84 customer contact centers and the intelligent routing, screen pop and FAX: 770-752-4545 iFLYTEK H other facilities in the U.S., Canada, call monitoring functionality provided www.gmvoices.com iFLY Mansion I HuangShan Road No. 616 B Latin America, Europe, the Middle by Envox’s award-winning CTI soft- Booth 307 I East, and Asia, and our global head- ware. Envox’s development tools and Hefei, Anhui 230009 China T

GM Voices delivers professional O quarters in Cincinnati, Ohio. For more software platforms are also used to recorded voices for all applications PH: +86 551 5397815 information, visit our website. streamline enterprisewide communi- and languages with consistent quality FAX: +86 551 5331801 R

cations through unified communica- at economical rates. On average, we www.iflytek.com/english D

tions solutions and to generate new I

deliver over 40,000 audio files a month R revenue streams for service providers. Booth 523 in over 70 languages. Combined with E iFLYTEK is the leading speech technol- our persona development, we provide C ogy solution provider in China. We have

a consistent voice brand throughout T First Data Voice Services a 70% market share in the Chinese your enterprise. O CRM Magazine 6200 S. Quebec St. speech technology industry. The com- Greenwood Village, CO 80111 pany owns speech synthesis, speech R

494 8th Avenue, 8th Floor Y New York, NY 10001 PH: 303-967-5698 Gold Systems recognition, oral language assessment PH: 212-251-0608 FAX: 303-967-5678 4840 Pearl East Circle and many other global leading technolo- FAX: 212-779-1152 www.firstdata.com Suite #106 gies. We can offer Chinese/English lan- www.destinationcrm2008.com Boulder, CO 80301 guage solutions for the desktop, plat- Booth 411 form and embedded application envi- PH: 800-988-7798 Booth 317 First Data powers the global economy ronment. Appointed by the Ministry of FAX: 303-447-0814 CRM magazine incorporates vital infor- by making it easy, fast and secure for Information Industry of China, iFLYTEK www.goldsys.com mation that will help you benefit from people and businesses around the is now leading the work to set up the the experience of other companies world to buy goods and services using Booth 410 Chinese speech technology standard. and ensure that your company virtually any form of payment. First Gold Systems develops and deploys becomes a CRM success story. The Data's Voice Services group powers voice-driven communications solu- magazine is FREE to qualified sub- automated customer self-service by tions utilizing IVR, text-to-speech and Inference Communications scribers at the Web site. developing and managing high-usage speech recognition. Our offerings Unit 1 inbound and outbound voice applica- include implementing unified commu- 174 High Street tions on our state-of-the-art hosted nications and messaging telephony Ashburton, Victoria 3147 Australia EMPIRIX platform. Put First Data's intelligence to interfaces, customized contact center PH: 03 8102 9352 20 Crosby Drive work for your business. solutions, and password management Bedford, MA 01730 FAX: 03 9571 6904 for telephone-based self-service. Gold www.inferencecommunications.com PH: 781-266-3200 Systems is a top-tier partner to leading FAX: 781-266-3201 speech and messaging vendors, Booth 522 www.empirix.com including Microsoft, Avaya, and Inference Communications is a special- Booth 313 Nuance. ist natural language speech recognition Empirix provides the only suite of auto- technology and solutions company. mated testing and monitoring solu- Inference's offerings include a wide tions designed specifically for contact range of pre-packaged speech solu- Genesys Telecommunications tions that can be customized, managed centers. By covering the lifecycle from Laboratories pre-deployment testing to customer and maintained by the customer via a experience monitoring, contact cen- 2001 Junipero Serra Blvd. simple web interface. All solutions are ters can identify problems before Daly City, CA 94014-3891 IBM underpinned by Inference's award win- agents and callers are affected. The PH: 650-466-1100 294 Route 100 ning grammatical inference technology. net result: reduced customer churn, FAX: 650-466-1260 Somers, NY 10589 lower operating costs and improved www.genesyslab.com PH: 914-766-1553 www.ibm.com customer loyalty. Booth 316 Genesys, an Alcatel-Lucent company, Booth 201 is 100% focused on software for cus- IBM Speech Solutions include a broad tomer interactions. Genesys provides range of software and services with a an integrated software suite for dynam- strong business partner ecosystem. Intervoice ic environments, automatically adjust- Offering IBM WebSphere enterprise 17811 Waterview Parkway ing to balance corporate resources, speech solutions, Embedded ViaVoice Dallas, TX 75252 marketing goals, and customer needs. for automotive telematics and mobile PH: 972-454-8000 With 3,300 enterprise customers in 80 devices, IBM Business Consulting FAX: 972-454-8707 Envox Worldwide countries, Genesys directs more than Services dedicated to transforming www.intervoice.com 2000 West Park Drive 100 million customer interactions every contact centers, and 40 years of Westborough, MA 01581 day. For more information, visit our research, including more than 2,400 Booth 604 PH: 508-898-2600 website. speech patents. Intervoice is a world leader in multime- FAX: 508-366-0009 dia communications, delivering natu- www.envox.com ral, intuitive ways for people to interact, Booth 511 Exhibit Hours transact and communicate. Our soft- Envox Worldwide is a leading global ware and professional services enable provider of IP-based voice self-service innovative voice portal, IP contact cen- and contact center solutions. The Monday, August 18 ...... 5:00 p.m. – 7:00 p.m. ter, hosted services and self-service company’s standards-based software Grand Opening Reception applications that empower millions of solutions dramatically reduce cus- people daily. More than 5,000 cus- tomer service costs, optimize contact Tuesday, August 19 ...... 10:00 a.m. – 5:30 p.m. tomers in 80 countries rely on center performance and improve cus- Wednesday, August 20 ...... 10:00 a.m. – 2:00 p.m. Intervoice.

WWW.SPEECHTEK.COM 43 Inventive Labs technologies enhance customer satis- call center technology solutions and Nortel Networks faction and greatly reduce costs. professional services. MicroAutomation 4955 E. Preserve Ct. 4000 Veterans Memorial Highway has designed, developed and imple- Greenwood Village, CO 80121 Bohemia, NY 11716 mented speech-enhanced solutions

Y PH: 866-923-5290 LumenVox PH: 631-285-2101 worldwide in a variety of industries such R www.voiceelements.com www.nortel.com/selfservice 3615 Kearny Villa Road as healthcare and government. O Booth 303 Suite 202 MicroAutomation’s mission is to reduce Booth 801 T Inventive Labs is a leading provider of San Diego, CA 92123 costs and enhance caller experience Nortel's Media Processing Server C royalty-free telephony application tools PH: 858-707-7700 voice portal products provide speech

E through call center automation. FAX: 858-707-7072 driven, converged, multimodal and R and solutions. The company’s soft- I ware products and related services www.lumenvox.com multimedia carrier quality to the enter- D

NeoSpeech include Voice Elements, a .NET toolkit Booth 303 prise, hosted or carrier customers. R empowering businesses to rapidly LumenVox provides powerful and 1290 Oakmead Parkway MPS is cost-effective, resilient, and secure; open-standard based, using O develop and deploy speech solutions. affordable speech recognition soft- Suite 318

T Inventive Labs is headquartered in VoIP/SIP and/or TDM to the network.

I ware together with a committed focus Sunnyvale, CA 94085 Greenwood Village, Colo. Experience the expertise of our exten- B on excellent customer service. This PH: 408-914-2710 I sive professional services organization year at SpeechTEK, LumenVox will FAX: 408-914-2710 H and take advantage of SOA, SCE, and highlight five partners who either use www.neospeech.com X our award-winning Speech Engine or VXML/CCXML. E provide complementary technology: Booth 311 Aculab, Cepstral, Inventive Labs, NeoSpeech offers superior text-to- Nuance Prosodie Interactive, and VClarity. speech software engines and applica- IQ Services 1 Wayside Rd. tions that deliver the most natural- 6601 Lyndale Avenue South Burlington, MA 01803 sounding synthesized voice. Our TTS Suite 330 PH: 781-565-5000 languages include U.S. English, Minneapolis, MN 55423 FAX: 617-565-5001 Spanish, Japanese, Mandarin Chinese, PH: 612-243-6700 www.nuance.com FAX: 612-869-6200 and Korean. Booth 501 www.iq-services.com Nuance (NASDAQ: NUAN) is in the Booth 408 Message Technologies Inc. Nexidia business of helping companies better IQ Services empowers companies to 1995 North Park Place 3565 Piedmont Road support, communicate with and deliver the best possible experience to Meridian - 5th Floor Building Two, Suite 400 understand their customers. It does their customers through flexible man- Atlanta, GA 30339 Atlanta, GA 30305 this through a set of interaction solu- aged testing services for contact cen- PH: 770-240-8000 PH: 404-495-7220 tions including Inbound Messaging, ter and communication solutions. Outbound Messaging, Analytics and FAX: 770-240-7474 FAX: 404-495-7221 From design to production, IQ Productivity. These solutions help www.messagetech.com Services delivers confidence about www.nexidia.com drive an improved customer experi- solution performance through com- Booth 508 Booth 803/805 ence, better business performance, prehensive investment protection and Message Technologies, Inc. (MTI) Nexidia is the market-leading provider and increased employee productivity test planning; performance and load helps companies improve business of highly scalable and accurate rich testing; application feature testing; performance using IVR and speech media search and speech analytics Nü Echo, Inc. availability monitoring; and other test technologies. Since 1982, more than software. Nexidia's patented phonetic services. 2,000 companies have counted on us search engine (PSE) enables contact 460 Sainte-Catherine W, Suite 424 for affordable, best-of-breed IVR and centers to gain valuable business intel- Montréal, QC H3B 1A7 Canada speech application design, develop- ligence from their recorded customer- PH: 514-861-ECHO ment and hosting solutions. MTI also agent interactions by analyzing call FAX: 514-861-1676 partners with leading speech applica- www.NuEcho.com tion developers to provide a reliable traffic, customer experience and busi- hosted infrastructure to support their ness opportunities. For more informa- Booth 413 Loquendo customer solutions. We have broad tion, please visit our website. Nü Echo develops the highest per- 745 Fifth Avenue, 27th Floor capabilities, configurable applications, formance speech applications in the New York, NY 10151 industry. Building on more than a custom IVR solutions, best of breed NICE Systems PH: 212-310-9075 platforms and a highly available host- decade of experience in speech appli- FAX: 212-310-9001 ed infrastructure. To learn more about 8 Hapenina Street cations development and a full suite of www.loquendo.com us, please visit our website. PO Box 690 application development, tuning, and testing tools, Nü Echo is uniquely Booth 509 Ra'anana, 43107 Israel PH: 972 9 775-3777 equipped to deliver applications that Winner of “Market Leader – Best MicroAutomation simply work better. FAX: 972 9 743-4282 Speech Engine” at the 2007 Speech 10550 Linden Lake Plaza, 2nd Floor Industry Awards, Loquendo is the www.nice.com Manassas VA 20109 leading innovator in speech recogni- Performance Technology PH: 1-800-817-2771 Booth 614 tion and speech synthesis, providing a Partners (PTP) FAX: 1-703-366-3850 NICE Systems (NASDAQ: NICE) is the complete range of technologies for leading provider of solutions that deliv- 11246 Gold Express Drive, Suite 101 server, embedded and desktop solu- www.microautomation.com er strategic and operational insights to Gold River, CA 95670 tions. Currently available in 23 lan- Booth # 314 the enterprises organizations by PH: 916-791-8263 guages with 57 voices, and constantly MicroAutomation is a leading systems recording and analyzing telephone FAX: 916-266-8708 growing, Loquendo’s best-in-breed integrator offering an extensive range of calls; website- email- and chat- con- www.performtechnology.com tent; customer transactions; and radio Booth 412 Email Stations communications. NICE is also the PTP's experienced team adds value to leading provider of Interaction your contact center by providing fea- Analytics Business Solution, designed ture-rich applications and complete sys- Email stations will be open in the Exhibit Hall during to improve contact center operational tem integration. We have expertise in regular exhibit hours for attendees to check their e-mail. efficiency and drive strategic enter- contact center technologies and their prise initiatives. integration into your mission critical

44 AUGUST 18–20, 2008 NEW YORK MARRIOTT MARQUIS, NEW YORK CITY EXHIBITOR DIRECTORY applications. Our professional services from recorded voice conversations to improve superior end-to-end service Vicorp range from project management, VUI discover customer needs and market for their customers. With more than 2 design, technical architecture design, trends and analyze calls for script decades of pioneering leadership and Wexham Springs development, system integration, to adherence, achievement of KPIs, root thousands of solutions deployed glob- Framewood Road E testing, tuning, and maintenance. cause and fraud detection. Leveraging ally, Syntellect is a premier provider of Wexham, Buckinghamshire SL3 6PJ X

SERTAINTY with your recording solu- enterprise-class contact center solu- U.K. H

tion helps to reduce call volumes, tions for the utilities, financial services, PH: +44 1753 660 500 I Persay Ltd. improve the customer experience, and B government, high- technology, help FAX: +44 1753 660 501 I 22 Zarhin St. increase revenue opportunities. desk, consumer prod ucts and health- www.vicorp.com T P.O. Box 4080 care industries. O

Ra'anana, 43665 Israel Booth 306 R

Vicorp makes it easy for organizations

PH: (972) 3 767 8666 D Tu Vo x to deliver superior services over the

+1-866-2398-515 I FAX: (972) 3 767 8661 19050 Pruneridge Avenue phone while significantly reducing R Suite 150 cost. Our experienced team of experts E

www.persay.com C Cupertino, CA 95014 and our smart technology can add Booth 321 T SpeechCycle PH: 408-625-1700 value to all organizations looking to Persay is a leading provider of O 26 Broadway, 11th Floor FAX: 408-625-1770 improve customer interactions and advanced biometric speaker verifica- R New York, NY 10004 www.tuvox.com gain competitive advantage. For more tion products with a global network of information visit our website. Y partners and system integrators. PH: 646-792-2710 Booth 409 Utilizing the biometric power of voice FAX: 646-792-2722 Since 2000, TuVox has become synony- to verify a speaker's identity, Persay's www.speechcycle.com mous with superior caller experience. Vocal Laboratories, Inc. products are used by leading financial, Booth 513 TuVox customers achieve improved 8421 Wayzata Blvd telecommunications, enterprise and SpeechCycle Rich Phone Applications customer retention and increased rev- Suite 260 law enforcement organizations. enue through innovative combinations deliver the intelligent fusion of natural Golden Valley, MN 55426 of technology and agent service. TuVox language dialogs with enterprise back- PH: 952-941-6580 end systems to create an immersive leverages customer investments in next- Prosodie FAX: 952-941-0819 caller experience. RPAs represent the generation technology with a choice of 855 SW 78th Avenue, Suite 100 future of enterprise and mass con- hosted, on-premise, and hybrid deploy- www.vocalabs.com Plantation, FL 33324 sumer speech applications, connecting ments, with self-managed and vendor- Booth 507 PH: 866-776-7634 web and other transactional systems managed service options available. VocaLabs solves the need for cost- www.ivrinc.com with a highly scalable, on-demand envi- effective and statistically accurate Booth 303 ronment that enables rapid time-to- VClarity feedback about client care speech value for phone-based initiatives. recognition, IVR, human agent, and Prosodie Interactive has been an 17581 Irvine Boulevard industry leading IVR service provider tone applications. Our clients are end- Suite # 106 for more than 20 years. Prosodie helps user enterprises, call center consult- Tustin, CA 92780 to increase efficiency and customer ants, application providers and equip- experience for contact centers by PH: 714-333-9465 ment vendors. We have the unique automating applications such as: www.vclarity.com ability to evaluate a customer contact inbound call routing/steering, branch/ Booth 303 application at any stage of develop- dealer locator, service & technical sup- Speech Technology Magazine VClarity provides enterprise voice solu- ment, from prototype to live in the field port, outbound marketing dialing with tions and a VoiceXML IVR platform for using a large and demographically 2628 Wilhite Court, Building Four hot transfer, virtual predictive dialing voice-automated self-service applica- diverse pool of panelists, or via Suite 100 and much more. tions, where customers can easily Express Feedback, a proprietary serv- Lexington, KY 40403 manage their applications and devel- ice that matches live caller feedback PH: 877-993-9797 Resolvity, Inc. op, test, deploy, and host voice-auto- with call recordings from the same FAX: 859-278-7364 mated solutions on a highly reliable individuals. 1333 Corporate Drive, Suite 200 www.speechtechmag.com and flexible voice foundation frame- Irving, TX 75038 Booth 217 work and LumenVox´s multi-lingual PH: 972-518-0863 Speech Technology magazine is rec- Automatic Speech Recognition (ASR) FAX: 972-518-0959 ognized worldwide as the leading and Text To Speech (TTS) provided by www.resolvity.com source of news, information, and Cepstral and Neospeech. Booth 617 analysis relating to the speech technol- Resolvity’s intelligent speech solutions ogy industry. It also provides addition- Verint Witness help companies automate a wide vari- al sources of news, information, and Actionable Solutions VoiceObjects, Inc. ety of customer support calls, ranging analysis through online communities 1875 South Grant St., Suite 720 330 South Service Road from natural language call-routing to at www.speechtechmag.com and San Mateo, CA 94402 www. speechtek.com as well as with Melville, NY 11747 more complex and interactive techni- PH: 650-288-0299 cal and product support applications. opt-in electronic distribution networks, PH: 631-962-9864 FAX: 650-525-9414 The company has expertise in high- STM eWeekly and SpeechSource. FAX: 631-962-9085 tech, telecom, cable MSO, healthcare Subscribe FREE today at www.speech www.verint.com www.voiceobjects.com techmag.com/subscribe. and other markets. Both hosted and Booth 310 Booth 800 on-premise solutions are available. Verint Witness Actionable Solutions VoiceObjects is redefining over-the- Syntellect, Inc. offers Impact 360, an analytics-driven phone customer service for global enterprises and carriers through adap- SER Solutions, Inc. 16610 N Black Canyon Hwy workforce optimization solution that tive, cost-effective self-service phone Suite 100 brings together quality monitoring, 45925 Horseshoe Dr., Suite 150 portals. VoiceObjects enables organi- Phoenix, AZ 85053 TDM and IP recording, speech and Dulles, VA 20166 zations to personalize each caller's PH: 800-788-9733 data analytics, workforce management, PH: 703-948-5500 performance management, elearning, experience, to integrate phone self- FAX: 602-789-2768 FAX: 703-430-7738 and customer feedback surveys into a service into comprehensive customer www.syntellect.com www.ser.com unified solution for analyzing customer experience strategies, and to manage Booth 322 Booth 510 interactions, improving workforce per- the complexity of the world's most SER's speech analytics solution, SER- At Syntellect, we help our customers formance, and optimizing enterprise sophisticated phone applications. For TAINTY, extracts valuable information create, maintain and continuously customer service processes. more information visit our website.

WWW.SPEECHTEK.COM 45 EXHIBITOR DIRECTORY

VoiceVerified, Inc. first solution that provides 100% appli- downloaded more than 20,000 times Center Applications. IVR, Speech cation testing and analysis. No other since its launch in 2006. Its standards- Recognition, PBX, ACD, Message-on- 110 Union Square Dr. testing methodologies or products based IVR hosting platform is the hold and websites are all systems New Hope, PA 18938 provide enterprises the validation that largest in the world and the only one Walsh Media can assist in enhancing Y PH: 215-862-7800 deployed applications are error-free. backed by a 100% uptime guarantee. R the customer experience. Since 1986, FAX: 215-402-1070 For more information, call or visit our

O we have supported the Fortune 1000 www.voiceverified.com website. T Voxeo and companies across the continent Booth 417 C 189 S. Orange Ave., #2050 to improve the way they communicate E VoiceVerified is a pioneer in the voice Voxify with their callers.

R biometric industry, providing verifica- Orlando, FL 32801 I tion services using existing telephony PH: 407-318-1217 1151 Marina Village Parkway D Alameda, CA 94502 infrastructures. VoiceVerified’s Point FAX: 407-835-0072

R Service Provider (PSP) on-demand www.voxeo.com PH: 510-545-5000 FAX: 510-545-5055 O hosted highly accurate voice verifica- Booth 804 T tion service enables call centers and www.voxify.com I Voxeo makes speech applications

B merchants to secure consumer not Booth 505

I easy to build and deploy. The compa- present (CNP) and other remote trans- Voxify enables contact centers to offer

H ny's common sense approach isn't actions, reduce costs, improve the rocket science: Offer a great platform. self-service for customer phone calls. West Interactive X customer experience, promote brand Automated Agents are speech appli-

E Make it exceptionally easy for people loyalty, and protect privacy. The 11808 Miracle Hills Drive to try, buy and use. Provide amazing cations that offer the lowest risk to Hosted Service enables rapid deploy- Omaha, NE 68154 support. This approach has made deployment and a rapid return on ment to enhance the speed of con- PH: 800-841-9000 Voxeo a leading choice among devel- investment. Built from industry-specific sumer adoption. FAX: 402-963-1602 opers and the answer for companies templates and patented conversation- that believe powerful telephony solu- al models, Automated Agents allow www.westinteractive.com Voiyager tions should be simple and affordable. callers to speak naturally resulting in the best customer service. Booth 502 16610 N. Black Canyon Hwy. More than 30,000 developers and West Interactive is a leading provider Suite 100 enterprise customers — including half of hosted and managed automated of the Fortune 100 — use Voxeo's Phoenix, AZ 85053 Walsh Media, Inc. customer contact solutions, custom Prophecy Platform and Hosting to PH: 800-788-9733 2100 Clearwater Dr., #201 speech applications, automated notifi- deliver innovative speech-driven IVR FAX: 602-789-2768 Oakbrook, IL 60523 cations, customer surveys, network- applications, outbound notifications, PH: 630-574-8008 www.voiyager.com sophisticated call routing and more. based call routing and analytic servic- FAX: 630-574-8118 Booth 514 Anyone can immediately download es. We make self-service a positive www.walshmedia.com Voiyager is a revolutionary technology the Prophecy Platform or sign up to customer experience that drives cus- that completely redefines how use Voxeo's hosted developer portal Booth 515 tomer satisfaction, reduces costs and VoiceXML applications are developed, for free at www.voxeo.com/free. Walsh Media, Inc., provides profes- generates revenue. For more informa- tested and deployed. Voiyager is the Voxeo's Prophecy Platform has been sional voice solutions for Contact tion, please call or visit our website.

COMPANY WORKSHOPS

Company workshops are open to all attendees and are located in the Manhattan Ballroom.

TUESDAY, AUGUST 19 WEDNESDAY, AUGUST 20

Convergys 10:00 a.m. – 12:00 p.m. IBM 1:30 p.m. – 3:30 p.m. w West Interactive 10:00 a.m. – 12:00 p.m. w w Making It Easy to Get What You Want Unified Communications — A Strategic Steven Cawn, IBM Speech Sales Leader Approach to Driving Greater Value This workshop will provide a detailed review of Exhibit Hours Rhonda Gibler, Senior Vice President, IBM’s Free Form Commands speech solutions, West Interactive which are an extension of IBM’s WebSphere Voice Unified communications enables relevant transac- Server and Embedded ViaVoice family of speech Monday, August 18 tions based on end-user need to drive improved products. This offering enables customers and 5:00 p.m. – 7:00 p.m. revenue and loyalty with maximum efficiency. Deliv- solution providers to implement a robust natural Grand Opening Reception ering on the right blended solution differentiates language call routing solution. IBM speech experts your brand by providing the greatest value to the will discuss application development and the avail- end user and great market share for your company. able development tools, implementation consider- Tuesday, August 19 Join West Interactive and learn how a unified com- ations, and tuning of Free Form Command appli- 10:00 a.m. – 5:30 p.m. munications strategy will improve your brand while cations. Both of IBM’s speech families (WebSphere reducing costs. Voice Server and Embedded ViaVoice) will be cov- ered, providing the attendees insight into natural Wednesday, August 20 language solutions for almost every speech appli- 10:00 a.m. – 2:00 p.m. cation in use today.

46 AUGUST 18–20, 2008 NEW YORK MARRIOTT MARQUIS, NEW YORK CITY AD AD