ISCA Archive Eurospeech 1999
ISCA Archive Eurospeech 1999

An asynchronous virtual meeting system for bi-directional speech dialog

Takuya Nishimoto, Hidehiro Yuki, Takehiko Kawahara, Yasuhisa Niimi

Voice-mail and video-mail systems for the Internet are becoming very popular, because the speech and video compression technology is greatly improved in these days. Asynchronous voice-mail system, however, does not seem attractive because it's not designed for discussions. We propose a network-based voice conference system which enables asynchronous and bi-directional discussions. The system displays the voice messages as the threaded written words. The user can manipulate voice messages just as if they are text messages. If the participant wants to quote or annotate to a message, the user has only to play the sound and barge into the message while it is playing. Such user interface increases the usefulness of the voice-mail system. We performed a preliminary experiment to evaluate the proposed conversation method. The voice messages are transcribed manually, because our system is not integrated with the speech recognition at present. Using the system, asynchronous voice conversations were realized eficiently. AVM is an application platform for the many kinds of speech processing, such as spontaneous speech recognition and automatic dialog tagging. The applications of the system include the entertainment systems, such as interactive radio dramas, or asynchronous multi-user game using speech dialog.