5th International Conference on Spoken Language Processing

Sydney, Australia
November 30 - December 4, 1998

Interactive Listening to Structured Speech Content on the Internet

Makoto J. Hirayama, Taro Sugahara, Zhiyong Peng, Junichi Yamazaki

Hewlett-Packard Laboratories Japan, Japan

Interactive information browsing of World Wide Web is a key application of the Internet and visual web browsers are widely used to access information. However, visual web browsing is not suitable in some circumstances such as in mobile environment. Therefore, we propose interactive listening to structured speech content for accessing information. Our proposed model of the interactive information listening services is that structured audio contents (HyperAudio) in a HyperAudio server are listened using a HyperAudio player whose appearance is similar to a portable radio. Unlike radio broadcasting programs, the HyperAudio contents have logical structures and hyperlinks so that listeners can listen desired information interactively. To put such logical structures into audio, a simple markup language was used. A prototype system of the HyperAudio server and players was implemented to test and evaluate feasibility and usability of the HyperAudio architecture.

Full Paper

Bibliographic reference.  Hirayama, Makoto J. / Sugahara, Taro / Peng, Zhiyong / Yamazaki, Junichi (1998): "Interactive listening to structured speech content on the internet", In ICSLP-1998, paper 1058.