Designed for attendees new to the speech technology arena, this tutorial provides an overview of today’s key speech technologies. What are the major differences between dictation and conversational recognition engines? Should you use statistical language models (SLMs) or speech grammars? Should you use concatenative or parameter-based speech synthesis engines? Do speaker identification and verification really work? Is there a need for touchtone recognition in interactive voice response systems? Who drives the speech dialogue? The user, the computer, or both? Where can natural language processing technologies such as natural language recognition, machine translation, response generation, and summarization be used today? Can speech technologies be embedded in mobile devices, or do they require a backend server? See how speech is being integrated with other modes. Untangle the voice standards alphabet: VoiceXML, SSML, SRGS, CCXML, PLS, and SCXML, and more.

Jump-start your knowledge in the field of voice user interface design. This session is designed to quickly get those new to speech and VUI design up-to-speed to make the most of the Customer Experience and Principles of Voice Interaction Design tracks. This tutorial will illustrate why VUI design is the make-or-break factor for speech applications and how to make smart design decisions from Day 1. Learn how to encourage customers to accept and use speech automation by focusing on the perceptions and reactions of end users throughout the design process. This tutorial will cover the basics in VUI design — the current and future state of technology (including multimodality); speech project methodology; design principles; and rules for efficient, no-nonsense call flows and evaluation techniques — so you can learn what works and what real customers think.

The proliferation of the internet and digital applications has resulted in increased digital identity theft. This session will cover the basics of digital signal processing and speaker identification, how noise filters work, the effectiveness of voice verification methods, drawing a balance between security and convenience, and the basics of voice print representation. This session will include a live demonstration of speaker voice registration and verification.

There are a variety of choices for speech recognition engines, especially for languages other than English. Although many companies are comfortable performing business case analysis across speech recognition engine vendors, it is often difficult to know how to compare accuracy. In this handson, in-depth course you will learn how to select an appropriate evaluation data set, why transcription is important and how to do it properly, how to perform the evaluation, and how to analyze and interpret the results.

You’ve done everything right in designing and deploying your IVR. Now what? Gone are the days — if they ever really existed — of putting an IVR in a closet and letting it run for years without thinking about it. An IVR needs to be constantly monitored for several reasons: to make sure it’s doing what it’s supposed to; to ensure the customer experience is staying where you want it; to increase self-service with time; and to change as the business changes. We’ll look at several concrete examples of how we’ve done exactly these things at Travelocity. The specific examples coupled with general good practices for ongoing IVR maintenance and monitoring should leave you with lots of ideas of what to look at in your own systems. We’ll look at reporting, A/B testing, and communication between the IVR team and the call center.

Without well-constrained grammars and lexicons to support it, a great design isn’t worth the paper it’s printed on. What types of concerns should be taken into consideration when designing complex grammars? What are some characteristics of a complex grammar task? What grammar features can be leveraged to optimize recognition? This session will begin with a very high level refresher on grxml grammar structure, and then delve into advanced topics in how to optimize recognition for a variety of complex tasks. We’ll discuss what makes these tasks complex and, in each case, some “how-to” methodologies for optimizing recognition given the task.

Methodologies for optimizing speech applications have grown and evolved with the industry. Just as holistic views of the application replaced narrow efforts focused only on the recognition engine, the newest approaches step back even farther from the engine and examine caller success from dialing to hang-up. This course will teach you how to manage a successful optimization project to get maximum value from your speech applications. Topics include setting optimization objectives, understanding caller behavior and intent, measuring dissatisfiers and inefficiencies, and identifying the highest-impact improvement opportunities. Using actual case
studies and class exercises, the course will cover a framework and methodology that you can use to manage both in-house and vendor/partner optimization projects.

This is a practical, hands-on session where the audience will learn about developing open-standards-based mobile applications that are multimodal. Participants can download for free the open-standards (Eclipse)-based
authoring platform to develop multimodal applications that run on all popular smartphone devices. At the end of the session, participants will develop greater appreciation for development process for multimodal enabling webapplications using W3C SCXML, InkML, and speech technologies.

There has been a lot of talk recently about the concept of VUI standards and whether or not they are necessary to help designers do the right things for callers. In a November Speech Technology article about VUI standards, several designers (including this presenter) agreed that VUI standards are not the final answer. For this presentation, we will build on the basic assumption that most designers already know intuitively what’s right for callers, and that the real difficulty is in convincing business decision makers (the customer) to employ proper design strategies. This talk will come in two parts: Part one will provide some rules that decision makers should take into account when designing an IVR, with descriptions that can be used to convince their customers to follow these rules. Part two will offer techniques and arguments to help convince customers to make sound design decisions. We will also explore tools, such as the use of data and reporting, as well as effective  caller survey techniques, which can be used as tools to convince decision makers to try our methods and evaluate the results before ruling against us.

The design and deployment of a bilingual speech application in the U.S. presents many challenges: defining business goals and user needs; incorporating cultural, political, and linguistic knowledge; and handling technical aspects such as special characters and language-specific acoustic models. This workshop is designed to provide a fun collaborative role-play setting for exploring some of the key challenges encountered when deploying an English/Spanish application for the U.S. Participants from various backgrounds and levels of expertise will complement each other, as teams work toward solving a series of mini-challenges along the course of a simulated project life cycle. Beginners to bilingual applications will learn a breadth of new information, while veterans can augment their skills by sharing their own experience, taking on new roles, and exploring less-familiar project phases. Knowledge of Spanish is welcome but not required.