Designing The User Interface For Speech Recognition Applications

Monday, April 15

The purpose of this one-day workshop is to bring together researchers and practitioners to discuss the design of applications that rely on speech as the primary medium for communication between the user and the system. The goal is to increase participants' understanding of the issues that face designers of such systems, exchange ideas, and establish some common ground among the diverse groups involved in creating speech-based applications.

Topics for discussion will focus on three areas.

State of the Technology

Participants will explore the current state of recognition technology and its implications for interface design. This may include combining speech and natural language, integrating speaker verification, barge-in, and wordspotting, or using systems that adapt automatically.

Design Principles

The group will propose design principles, touching on techniques for creating conversational dialogs, making all the functionality of a system known to users, recovering from errors, using sound effects, and designing for multiple speech recognizers.

Usability Engineering

Participants will share their experiences on design methodologies for speech interfaces, including usability engineering strategies, the use of Wizard-of-Oz techniques, and experience with multi-lingual, international, and regional design.

Prior to the start of the workshop, participants will be expected to read all the position papers as well as experiment with several teleTel-based speech applications. A subset of workshop participants will also be asked to give a short oral presentation of their position paper.

This one-day tutorial is limited to 16 people.

Contact

Amir Mane
AT&T Bell Laboratories, 1J-325
101 Crawfords Corner Road
Holmdel, NJ 07733 USA
E-mail: amir.mane@att.com
Tel: +1-908-949-7049
Fax: +1-908-949-8569