Hear about Google’s vision for speech technology. Following months of development and speculation, Google recently released its first speech application, GOOG411. Mike Cohen will describe Google’s experience with GOOG411, discuss Google’s general philosophy and approach to speech services, and review some of the lessons learned thus far. 

Speech-enabled applications in the call center make a myriad of self-service options available to the end user. However, the idea of “if we build it, they will come” has proven false for many organizations deploying speech. Why are some speech applications well-tolerated and some avoided at all costs? What are the factors that cause users to abandon automated systems in favor of live agents? Experts in this session consider speech technology as a part of an overall self-service strategy. Learn techniques for strategic planning, data collection, and analysis that will help create self-service applications that end users actually want to use.

Want to know why callers are abandoning automated systems? Ask them. At the main transfer points, callers are asked to state the reason for their call. By matching what callers told us to the self-service options they chose, we discovered key application performance issues that led to changes in design, verbiage, and the introduction of “supercharged” grammars.

How do you grow customer relationships when so many calls are automated? How do you convert service into sales without annoying customers? Explore best practices for promoting organic growth and customer loyalty using speech-automated cross-selling and up-selling. Learn specific techniques for designing timely and relevant offers.

This session looks at defining a self-service strategy from a holistic perspective: externally from the consumer’s perspective and internally from the business and technical perspective. Learn how to find the pitfalls in the design of a self-service strategy, how to meet the objectives of consumers and call center business and technical teams, and how to use technology to meet the objectives of self-service.

Usability is widely recognized as a measure of the quality of a voice user interface, and usability testing is a must-have in all VUI design projects. But does usability tell the whole story? These experts agree that excellent speech applications are more than just easy-to-use. In this session, hear cutting-edge ideas about what to measure beyond usability and how it can improve your speech application.

You’ve heard it over and over again, you’ve tested for it, you’ve thought about it, you’ve designed your application to get it. But when is usability not enough? This session focuses on usability plus other outcomes you need to consider when you’re designing your next application.

This presentation offers an overview of best practices for a) how to define speech application success criteria; b) how to instrument a speech application so that the desired numbers can be measured; and c) how success criteria influence the application design. Examples will be shown using several case studies in which different success criteria have greatly influenced the final application.

More of the world is going mobile and a new generation, the mobile generation, is using their wireless phones for more than just voice communication. The recent introduction of the iPhone is one such example. In this lunch presentation, Beatriz Infante, CEO of VoiceObjects, will introduce you to this mobile generation and show the next generation of applications they expect, not just on the iPhone but on every phone.

How do you know if your speech application is living up to your objectives? Is the application meeting the goals you set when you started the project? You’ll only know the answer to these crucial questions if you establish success criteria, tied to specific metrics, before the project begins. In this session, learn how to develop rigorous, meaningful criteria that will allow ongoing evaluation and improvement of your speech applications.

What defines a successful speech application? The answer may involve many different elements. Some are measurable analytics, and other aspects are more qualitative, such as caller satisfaction and customer perception. Further, success criteria definitions are not the same for every application. The key challenge is to clearly identify early on the business goals from the customer perspective and use them to drive the definition of success metrics.

In this session we’ll examine key dimensions of success for speech applications: effectiveness, efficiency, utility, and attractiveness. We’ll examine some of the successful methods employed by customers for measuring these dimensions, including both databased measurement, as well as experiential measurement, through direct customer feedback. We’ll also review strategies for migrating from “the old metrics” to the new metrics.

In years past, the decision to deploy speech was all about cost reduction—companies used speech applications to offload tasks from more-expensive live agents. Many organizations are realizing that this simplistic model does not always work as advertised, and that speech can do more than just reduce costs. Experts in this session detail new ideas for maximizing the value of speech applications and using speech for more than cost savings.

To reach its full potential, speech must enable more than higher automation rates in traditional IVR applications. Triggered by maturing standards and a broadening audience of developers, a new movement of creative speech development is emerging as a significant market segment. Learn how the Web model of innovation has been applied to speech applications.

With 411 fees rising to $2 or more per call, 1-800-FREE411’s ad-supported free directory assistance has saved millions of consumers millions of dollars in needless charges. With higher success rates and lower costs than before, speech recognition is critical to 1-800-FREE411’s caller experience, making free 411 an exciting new media opportunity.

Sixty percent of calls fail to achieve productive results. Incessant routing by automated systems keeps callers longing for the good old days of talking to human agents. How can we make good use of automation without losing the personal touch that’s so important to customers? In this session, panelists consider all customer communications as one flow, fusing contact with live agents with automated processes. Attendees will learn from the panelists’ real-world experiences about how customer service organizations are using new technologies to bridge the human-automation divide.

Hear about Google’s vision for speech technology. Following months of development and speculation, Google recently released its first speech application, GOOG411. Mike Cohen will describe Google’s experience with GOOG411, discuss Google’s general philosophy and approach to speech services, and review some of the lessons learned thus far. 

A voice user interface is a balancing act between the goals of the business and those of the end users. These goals are often in conflict—businesses want to push more calls to self-service, users want total access to live agents—often to the detriment of the success of the application. When should user requirements win out? In what cases are business requirements more important? The experts in this session provide the knowledge you need to answer these questions.

Business owners have business goals, objectives, and requirements. Designers bring experience and advocate user needs throughout the design process. So how can we create outstanding experiences when objectives may seem to clash or customers have preconceptions about “how the system should work”? Explore some common challenges, understand the real issues behind resistance, and discover how to focus instead on successful systems.

By the time an application has the go-ahead from executives, requirements are driven by the business and not the caller. Learn how to find out who the caller really is and how to take several steps back to design for the true caller, so your application is actually used and liked. Business requirements are important, but it’s essential to find the right balance.

Surveys are an important method of getting opinion feedback from users of speech applications. At best, surveys provide quantifiable data that clarifies user opinions, but many do-it-yourself surveys do not achieve this result. In this session, you will learn how to craft surveys that deliver reliable, accurate data to improve the performance of your speech application. Attendees will gain a basic understanding of survey theory, methods, techniques, and analysis.

More of the world is going mobile and a new generation, the mobile generation, is using their wireless phones for more than just voice communication. The recent introduction of the iPhone is one such example. In this lunch presentation, Beatriz Infante, CEO of VoiceObjects, will introduce you to this mobile generation and show the next generation of applications they expect, not just on the iPhone but on every phone.

Prompts are at the heart of any VUI design. The embodiment of the sound and feel of the application, prompts convey both affective and informational content. In a very real sense, the usability of a speech application is largely determined by the quality of its prompts. Crafting effective prompts is a creative and scientific endeavor, requiring a diverse skill set. This expert VUI designer outlines an approach for writing, recording, coaching, and processing prompts to ensure the highest quality possible.

Come share your experience! Audience members will actively participate in this session, sharing insights and anecdotes on the do’s and don’ts of how to use speech automation to create a better end-to-end customer experience. End users don’t evaluate speech automation in isolation—they view it as part of an integrated customer service chain. So in designing the optimal VUI, it’s important to take into account what might happen before, during, and after the automated speech interaction in order to create a seamless customer experience.

Speech projects always involve multiple contributors, often with diverse backgrounds and differing levels of understanding of project goals and speech technology itself. The voice user interface designer often sits squarely in the middle of a group of project sponsors, developers, call center and telephony managers, and others who have a stake in the success of a speech project. In this session, experts suggest effective techniques for facilitating communication both within the team delivering the speech application and between the team and project sponsors.

This presentation stresses the importance of incremental and modular descriptions of system functionality for targeted and phased reviews and testing. This strategy ensures clarity, consistency, and maintainability beyond the project lifetime and eliminates the need for changes midproject, thus both managing customer expectations and protecting the service provider from ad-hoc requests.

Teaming is an essential, complicated, and stressful aspect of technology development. This session focuses on what makes a team function well, the most common teaming problems in speech projects, and ideas for troubleshooting to make your team highly effective!

Hear about Google’s vision for speech technology. Following months of development and speculation, Google recently released its first speech application, GOOG411. Mike Cohen will describe Google’s experience with GOOG411, discuss Google’s general philosophy and approach to speech services, and review some of the lessons learned thus far. 

Advances and improvement in core speech recognition technology are difficult to demonstrate, since accuracy is strongly dependent on application, particular speakers, background noise, and other variables. Beyond accuracy, speech recognition technology can be improved by better handling of complex or “natural” dialogs. Audio channels and speech platforms are important components of today’s speech applications. In this session, speakers explore the advances in core speech technology, audio channel processing, and speech platform integration and go behind the scenes of Vista to expose interesting aspects of the integration of speech technology.

Windows Speech Recognition (WSR) in Vista is a practical solution for speech-enabled access to Windows-based PCs for users who find keyboard and mouse interfaces to be less productive than they would like. The technology behind WSR includes automated personalization, the Microsoft Speech Recognizer, SAPI 5.3, the accessibility framework, the text services framework, and Windows Desktop Search all being employed to deliver the Windows Speech user experience.

There are two methods for compressing and transmitting digital speech for server based automatic speech recognition. Distributed Speech Recognition (DSR) schemes gained popularity in the late 1990s due to limited data channel bandwidth availability. The evolution of higher bandwidth channels and advances in voice compression now allow Network Speech Recognition (NSR) applications to achieve the speech recognition accuracy of DSR in similar bandwidth and provide additional benefits. This presentation compares voice-based NSR with features-based DSR recognition schemes.

Text-to-speech synthesis is getting better, more flexible, and is now used globally in a wide spectrum of speech applications. Advances in standards have improved text-to-speech quality. The Speech Synthesis Markup Language (SSML) provides a standard way to control speech synthesis and text processing parameters. The Pronunciation Lexicon Specification (PLS) is designed to enable interoperable specification of pronunciation information. This session reviews some much-needed clarifications about how text in multiple languages should be annotated and describes work being done to link SSML and PLS more seamlessly.

Many speech applications demonstrate the need to define the pronunciation of certain words (for instance proper names, locations, etc.) or to expand acronyms/abbreviations, both for ASR and TTS usage. This presentation describes the W3C PLS (Pronunciation Lexicon Specification) that defines lexicon documents to be referenced by SRGS grammars and SSML prompts.

In SSML, how do you mark tones, or use pinyin for pronunciation, or indicate a change in language but not a change in voice? Learn about the changes in SSML that provide improved support for Mandarin, Cantonese, Japanese, Hindi, and other world languages. This session also explains multi-language annotation and how to link with PLS.

More of the world is going mobile and a new generation, the mobile generation, is using their wireless phones for more than just voice communication. The recent introduction of the iPhone is one such example. In this lunch presentation, Beatriz Infante, CEO of VoiceObjects, will introduce you to this mobile generation and show the next generation of applications they expect, not just on the iPhone but on every phone.

The demand for natural language has reached an all-time high as directed dialog applications continue to be criticized for being inefficient and not flexible enough. There is little dispute that out-of-grammar handling is generally poor when an active grammar is large. In-grammar accuracy for extensive vocabularies has been achieved by using large amounts of speech data to extract statistical information to represent acoustical units. Likewise, statistical approaches have been applied to advance natural language understanding. Most recently, statistical approaches are being applied to voice interface design with the goal of improving user experience. This session reveals some exciting advances in natural language that will affect the future of the user experience.

For the current state-of-the art speech recognition systems, the in-grammar accuracy is quite good, especially for directed-dialog systems. However, due to the variability of how callers respond, a portion of the utterances are not covered by the grammar, i.e., they are out-of-grammar (OOG). OOGs affect the “perceived” accuracy of the system, and are one of the primary items addressed during tuning. This presentation discusses the concepts of “near OOGs,” “far OOGs,” and related concepts.

Today we are extending the data exploitation paradigm to voice user interface (VUI) design. Statistics and machine-learning sciences are now complementing the art of designing the best prompts and interaction strategies with the goal of optimizing automation and improving user experience. Using a few case studies, this presentation shows how to “experimentally” choose among competing VUI designs without disrupting the user experience while optimizing global indicators of performance.

Recent innovative integration of recognition and synthesis technology has led to the realization of fully automatic speech-to-speech translation. This session explores the latest techniques for implementing automated language translation and considers the technology behind the integration: how to manage out-of-grammar responses, the effects of using robust parsing versus SLMs, and incorporating an open source speech analytics solution called Unstructured Information Management Architecture (UIMA).

This presentation shows a distributed infrastructure for integration of third-party recognition, translation, and synthesis technologies into speech-to-speech system combinations. The infrastructure is built over the Unstructured Information Management Architecture (UIMA), an open-source framework for speech analytics. The Web infrastructure has successfully been used for the remote automatic evaluation of speech-to-speech systems on pan-European scale.

As automated language translation technology moves to statistically based computational methods, the timing is right to integrate language translation and speech recognition technologies. Case study examples and demos of existing integrated solutions will give the audience an overview of how to leverage speech applications across languages.

Voice search is perhaps the hottest topic in recent speech deployments. Analogous to searching the Web with text, voice search can encompass a number of services, including directory search and searches for specific information, such as news or sports scores. What are the requirements for achieving effective dialogs when searching by voice? How does dynamic content, such as location-based ads, fit into the voice-user interface? What other analogies are there between voice searching and Web searching? This session is a must for those interested in learning about the trends in voice search.

Voice search is very hard to do well when you consider the millions of different accents, behaviors, and speech patterns a software program would have to decipher. What is the best way to architect the solution so that it has the best chance of providing an effective consumer experience? What are the business considerations for making it work in the real world? In this presentation you will hear thoughts and learnings from the edge of the “voice search” frontier.

Voice search topics and Web content change all the time. How can an architect prepare the recognizer to recognize fundamentally new words and topics? With all of the activity on the Internet, are there any useful data sources for recognizer training? This presentation will explore how the Web and other data sources may be leveraged to keep your voice search solution current.

Hear about Google’s vision for speech technology. Following months of development and speculation, Google recently released its first speech application, GOOG411. Mike Cohen will describe Google’s experience with GOOG411, discuss Google’s general philosophy and approach to speech services, and review some of the lessons learned thus far.