The market for speech technology topped the $1 billion mark in 2006, up 100% from 2 years ago. As adoption of speech systems increases in all parts of our lives, from our computers to our cars to customer relationships, it is clear that speech technology is at the tipping point. Hear from Malcolm Gladwell, who coined this phrase, about how he sees ideas, behaviors, and patterns moving through a population in much the same way as an epidemic. He posits that ideas can become contagious and reach a tipping point, just as a virus reaches critical mass. Learn what it takes to start a social epidemic and create change and how to push speech technology, and your businesses, toward the tipping point.

Outsourcing is a common practice and is often the best practice for costeffective, high-quality customer service and self-service. In this session, leading service providers and their customers introduce and describe the criteria that influenced their decision to outsource and their experiences with the team of application developers, carriers, and hosted services providers.

Nearly every service provider offers savings in capital expense and contact center operating costs. Increasingly, service providers differentiate themselves through partnering strategies, technology in use, application development and management resources, and support of multiple modalities of customer care and self-service.

This has been the year of the end user. With so much emphasis on the customer experience, being able to identify snags in self-service applications at a highly granular level and to make corrections dynamically is crucial. In this session, application framework providers describe how they support application development that meets customer service objectives.

In this session, the biggest stakeholders in global, outsourced communications discuss how the distributed, virtualized contact center will integrate speech automation with multimodal communications (blending live agents with IM, chat, and video messaging) and what this means for the future of customer service.

The market for speech technology topped the $1 billion mark in 2006, up 100% from 2 years ago. As adoption of speech systems increases in all parts of our lives, from our computers to our cars to customer relationships, it is clear that speech technology is at the tipping point. Hear from Malcolm Gladwell, who coined this phrase, about how he sees ideas, behaviors, and patterns moving through a population in much the same way as an epidemic. He posits that ideas can become contagious and reach a tipping point, just as a virus reaches critical mass. Learn what it takes to start a social epidemic and create change and how to push speech technology, and your businesses, toward the tipping point.

The world is getting smaller, and increasingly this is reflected in speech applications that include more than one language. Enabling access to a speech application in another language is not a simple matter of translating the prompts and switching the recognition algorithm to the other language. Experts in this session offer tips on adding a second language to an existing application and how best to approach the design of new multilingual applications.

What type of Spanish is the best to use for the Hispanic population in the U.S., who have backgrounds from more than 20 different countries. One solution that is not discussed often enough and that upsets language purists is mixing English and Spanish, or even creating hybrid words in “Spanglish.” This presentation addresses linguistic facts about Hispanics in the U.S. and the appropriateness of using mixed-language phrases and Spanglish words.

There are many challenges to adding a second language to your speech application. Hear useful advice in this presentation to help you ensure a smooth design and development process, including real-life, concrete examples drawn from practical experience working on various Spanish voice applications.

There is a growing need to design and develop speech-enabled, self-care applications that are bi- or multilingual (e.g., Spanglish, Hinglish etc.). In these design projects, it is important to understand cultural differences, analyze speaking styles, sentence constructs, etc. Learn about challenges and solutions for tailoring and designing these applications.

Recognition errors are a fact of life for speech systems. The job of a good voice user interface is to make the errors as inconsequential as possible to the end user. Learn how to handle confirmations to build user confidence without becoming a burden, and hear about new approaches to error-handling for natural language applications.

Confirmations are a critical piece to the VUI puzzle. This presentation investigates various confirmation strategies and illustrates successful phrasing techniques, including a natural-sounding way to confirm dynamic data. Attendees will expand their confirmation toolkit and learn how to improve caller satisfaction and decrease call duration in their own IVR applications.

Hear about proven strategies for building fault-tolerant VUIs for smarter speech applications. The speakers discuss the use of statistical language models, error strategies tailored to specific call paths, and the careful use of confirmations to minimize errors, improve performance, and increase overall customer satisfaction.

Speech applications used to be conceived of as one-size-fits-all interactions in which every user had the same experience. Increasingly, organizations are capitalizing on data they already know about customers to tailor the experience specifically for them. Whatever you know about the customer can be used to predict the likely reason for the call and present relevant information and services proactively. Learn techniques for creating personalized interactions that make calls seem efficient, productive, and surprisingly satisfying.

A great application is not only smart and engaging, but also relevant. This session looks at how you can draw on intelligence about a user to adapt and tailor the experience for him. Topics include streamlining a user interface and offering features proactively based on usage patterns and predictions of intent—and logistical challenges you’ll encounter. This approach to design is especially effective for mobile applications in which a phone identifies a single user.

This session shows how to create a personalized contact center application using Cisco Call Studio, speech recognition, VoiceXML, and Web or enterprise applications. Learn how the Cisco Customer Voice Portal and interactive voice response technology built into Cisco IOS work together to allow you to build and deliver self-service applications.

Speech applications are only as effective as the grammars they reference. As such, grammars should be constrained as tightly as possible, and grammar over-generation should be avoided at all costs. This session’s speakers discuss various methods available to help constrain grammars, as well as some grammar development pitfalls and the testing methodologies that will help avoid them.

The market for speech technology topped the $1 billion mark in 2006, up 100% from 2 years ago. As adoption of speech systems increases in all parts of our lives, from our computers to our cars to customer relationships, it is clear that speech technology is at the tipping point. Hear from Malcolm Gladwell, who coined this phrase, about how he sees ideas, behaviors, and patterns moving through a population in much the same way as an epidemic. He posits that ideas can become contagious and reach a tipping point, just as a virus reaches critical mass. Learn what it takes to start a social epidemic and create change and how to push speech technology, and your businesses, toward the tipping point.

As designers are urged to create ever-more sophisticated self-service applications, the pressure for evolving new techniques grows in importance. New Eclipse-based graphical tools oriented around the identification, definition, and reuse of hierarchical dialog patterns and novel nonlinear call flows assisted by agents are described in this dialog design session.

This presentation shows an Eclipse-based, graphical environment for developing speech applications that specifically addresses the problem of capturing and expressing recurring dialog patterns. This tool transforms the process of designing and implementing dialogs by specifically orienting the design process around the identification, definition, and reuse of hierarchical dialog patterns.

Most speech IVR applications are unable to skip utterances that they don’t understand. In contrast, live agents can gather information out of sequence or discern intent. By uniquely combining speech technology with humans, callers can have a more natural, free-flowing self-service experience because they are not locked into a rigid call flow.

This presentation describes the technology and applications of automatic speaker classification (e.g., age, gender, language, and emotions) in voice portals. It shows how dialog parameters should be adapted to achieve improved user acceptance in IVR systems. The first results from pilot implementations within Deutsche Telekom prove the feasibility and show advantages compared to conventional, non-adaptive systems.

The growing sophistication of VUI designs demands the incorporation of new technologies, including those borrowed from other disciplines. This session focuses on the novel application of artificial intelligence technology concurrently using a dialog engine and a problem-solving engine. It also illustrates the use of natural language to understand the semantics and context of any phrase being processed, making it much easier to develop the answers.

This presentation discusses how AI technologies may be used in voice self-service applications to separate the product support logic from the call flow logic, take advantage of the problem-solver’s knowledgebase to develop dialogs, improve speech recognition, create dynamic call flows, and provide an effective and efficient troubleshooting experience. Learn about strengths and weaknesses, rule-based systems, Bayesian inference, decision-trees, and knowledgebases.

Natural language understanding is an excellent augmentation to an existing or new VUI for better automated call completion and customer satisfaction. Hear about a solution that makes it easier to develop the natural language application after understanding the semantics and context of any phrase.

The availability of a robust 3G infrastructure throughout Europe and much of Asia has released pent-up customer demand to add live video to extend the utility of voice communications. This session illustrates how video menus, pictures of products, live video clips, and video commercials can be managed, as well as how sample speech/video-enabled self-service applications for universities, travel, retail, and home health can be developed. Also, the use of the Adobe Flash Player, a popular standard for delivering rich Web content to develop multimedia content, will be explained and illustrated.

Around the world (including in the U.S. in 2007), many consumers have access to live 2-way video. Learn how speech-enabled self-service applications can become even more effective by showing the caller a visual display or video at the same time that they use the speech application—menus, pictures of products, live video clips, and video commercials. See sample speech-enabled self-service and proactive contact/outbound applications for universities, travel, retail, and home healthcare, and hear about the process of application authoring.

Learn about extending the Adobe Flash Player with speech recognition. A lightweight, embedded VoiceXML browser (VoiceXML 2.0-compatible) is easy to control through XML protocol from Action Script to speech-enable existing or new Flash applications. The VoiceXML is controlled by a browser extension for the IE and Firefox browsers. The browser uses the IBM ViaVoice Embedded Engine for speech recognition. Hear the key aspects of the design and about the challenges faced during the implementation.

Recognition technology has matured to the point that recorded telephone-quality audio from unknown speakers can be accurately transcribed. Applications such as speech-enabled e-mail have become highly needed in the mobile environment because typing is not always practical when using hand-held devices. Recent applications of speech-to-text for searching and transcribing voice data will be illustrated for other applications, including medical data transcription and the near-real-time conversion of voice mail to text.

As speech transcription technology improves and evolves, more opportunities emerge for captioning broadcast media. This presentation outlines some remaining challenges, such as latencies and understandability. It also addresses issues in combining speech transcription with other natural language technologies, such as search, translation, and named entity detection.

Speech-to-text has steadily improved in accuracy during the past 2 decades, but the question remains: “Is it good enough?” The answer lies not in the technology, but in the applications. Using her experience with BBN’s STT engine, Marie Meteer describes how STT performance affects a variety of applications: where it works, where it fails, and where supporting technologies can make the difference.

The market for speech technology topped the $1 billion mark in 2006, up 100% from 2 years ago. As adoption of speech systems increases in all parts of our lives, from our computers to our cars to customer relationships, it is clear that speech technology is at the tipping point. Hear from Malcolm Gladwell, who coined this phrase, about how he sees ideas, behaviors, and patterns moving through a population in much the same way as an epidemic. He posits that ideas can become contagious and reach a tipping point, just as a virus reaches critical mass. Learn what it takes to start a social epidemic and create change and how to push speech technology, and your businesses, toward the tipping point.

This session shows two case studies that illustrate how speech technology can be used in emergency situations. The Italian police use call steering to determine if the call is an emergency, and fire fighters use advanced, multimodal, mobile services enabling on-the-move rescuers to communicate and to share structured multimodal information resources including audio, video, text, graphics, and location information.

Hear how the Dortmund Fire Fighters (Germany) are using an advanced, multimodal, mobile service for emergency teams to handle large-scale rescue operations. The speakers will share their experience about how innovative technologies allow workers on the move to communicate naturally and bi-directionally and to share structured multimodal information resources, including audio, video, text, graphics, and location information.

Natural language processing analyzes incoming calls to determine if any represent an emergency call. In this presentation, learn about the innovative techniques used by the Italian police.

Synthetic interviews enable callers to go beyond the picture or cultural object to better understand the event or concept it represents. For instance, callers may conduct synthetic interviews with individuals in photographs of historical events in a museum or objects in a cultural heritage site. Synthetic interviews represent a new way of interacting with objects that are usually only viewed. The first case study in this session describes a cell-phone application, and the second shows a multimodal application that contains a chat-bot. When applied to the world of advertising, synthetic agents will enable people to talk with advertisements in magazines, on billboards, and other advertising media.

Many museums around the world rent audio players to their visitors to provide automated tours delivering prerecorded information about their exhibits. One alternative to the rental paradigm is using the patron’s own “digital sound player,” i.e., their personal cell phone. A system-directed dialog using VoiceXML can allow patrons to obtain a wide range of information through a menu-driven hierarchy or by questioning a synthetic agent virtual tour guide, docent, or teacher.

PDAs with ad hoc, built-in information retrieval and auto-localization functionalities can help museum visitors more naturally than using traditional audio/visual prerecorded guides. This presentation shows a userfriendly, multimodal virtual-guide system called MAGA, which is usable on multiple mobile devices (e.g., PDAs, Smartphones). An information retrieval service is accessible through a spoken-language interaction or an auto-localization service.

Around the world, speech applications are being used to update account information and to transfer billions of dollars among accounts. Learn how two major banks have successfully deployed speech applications that revolutionized their operations. Speakers will describe solutions to problems encountered during development and deployment.

They said it couldn’t be done. In today’s high-tech environment as more and more older consumers interact with speech technology, businesses are trying to find ways to make the experience a pleasant one. Hear the latest research findings and how one organization is strategizing to find ways to engage older consumers.

The market for speech technology topped the $1 billion mark in 2006, up 100% from 2 years ago. As adoption of speech systems increases in all parts of our lives, from our computers to our cars to customer relationships, it is clear that speech technology is at the tipping point. Hear from Malcolm Gladwell, who coined this phrase, about how he sees ideas, behaviors, and patterns moving through a population in much the same way as an epidemic. He posits that ideas can become contagious and reach a tipping point, just as a virus reaches critical mass. Learn what it takes to start a social epidemic and create change and how to push speech technology, and your businesses, toward the tipping point.

Specifying grammars that cover the words spoken by users is one of the major keys to designing and implementing speech applications. Tuning grammars is a complex task and requires automated techniques to analyze large amounts of data. What techniques can designers and implementers use to build and tune grammars to handle unexpected events in caller responses such as background talk, side conversations, nonspeech events, restarts, and corrections? Can some of these techniques be automated?

Speech applications misrecognize many common events such as background talk, restarts, and corrections and treat them as valid responses with a high confidence score. This presentation discusses techniques for building robust grammars that handle these unexpected events and other reasons for frequent speech recognition errors and will also review experimental results using data from speech applications.

Tuning grammars for large telcos and automating directory assistance services is a complex task and requires automated techniques to go over large amounts of untranscribed data. The key technical challenges relate to the integration of large vocabulary recognition of naturally spoken requests, disambiguation, and search and to developing automated approaches to building and tuning these grammars. In this presentation. hear two experts describe their experiences with building and tuning these grammars.

Text-to-speech synthesis is used to render dynamic data, such as newsfeeds and e-mail, and to avoid costly, time-consuming prerecording. The use of speech synthesis will increase as functionality, including pronunciation accuracy, natural timbre, intonation, and expressive voices that supply so much of an application’s persona, improves. Special attention will be given in this session to the challenging requirements of non-English-languages and new extensions to the Speech Synthesis Markup Language (SSML).

Making a speech synthesis system for Chinese language has encountered many linguistic and phonetic problems that differ from Western languages. iFLYTEK has taken the language characteristics and constructed a Chinese TTS system with the help of the China national standard CSSML and the developing W3C SSML 1.1 specification. The 2008 Olympics and World Expo 2010 in Shanghai will boost the utilization of the Chinese TTS technology and the markup language for Chinese text.

Davide Bonardo demonstrates the many TTS features available today in TTS products, how to use them to obtain effective prompts, and how to optimize TTS use. He will illustrate how to get the most out of the tools available in order to refine prompts for use in the widest variety of contexts.

A major problem facing speech application developers is integration with the rest of an enterprise’s IT infrastructures. Learn how to integrate speech user interfaces over a network with the rest of an enterprise’s IT infrastructure using service-oriented architecture (SOA) technology and Web services. The benefits of this approach include keeping your application portable and maintaining a clear separation of your call flow and your business logic.

One shortcoming of many speech applications is the lack of integration with the rest of an enterprise’s IT infrastructures. Web services integration between speech applications and other enterprise IT assets can remedy this situation. This presentation discusses service-oriented architecture (SOA) technology and Web services and the application of Web services to call center and other speech-enabled environments. See an example of a speech application that fully leverages CRM, BI, and other IT assets via Web services.

This presentation discusses functionality beyond the VoiceXML/CCXML standards and suggests a standards-based Web service methodology. The benefits include keeping the application portable and a clear separation of the call flow and the business logic. Learn how Web services can help your speech application leverage other standards without comprising portability.

Even the most carefully planned deployments can go wrong. Experts in this session review lessons learned from years of customer deployments of applications with two to 20,000 ports. They discuss the features of Session Initiation Protocol (SIP), a lightweight, transport-independent protocol for messaging, and share tips on how to overcome the challenges of deploying a VoiceXML/CCXLM application into an SIP environment. Hear “R.J.’s Rules” for successful speech deployments, encapsulating lessons learned from processing more than a billion calls.