Shinsuke Sakai Media Archiving Laboratory Academic Center for Computing and Media Studies Kyoto University Yoshida-nihonmatsu-cho, Sakyo-ku, Kyoto, 606-8501, Japan office phone: +81-75-753-9058 fax: +81-75-753-9056 email:

I am interested in human language technology such as speech recogniton, understanding, synthesis, and various facets of natural language processing. I have been working on corpus-based speech synthesis these days including additive models aproach to F0 modeling. Recently I am interested in topics such as on-line adaptation, hands-free (distant talking) speech recognition, and use of prosodic information in speech recognition. I am also generally interested in: Application of statistical learning/machine learning techniques to speech and language technology, such as speech recognition, synthesis, machine translation. Speech recognition and understanding technology for a listening agent (or robot) that do not require close-talking microphone and push-to-talk button. Various acoustic modeling issues for speech recognition.
	f0 modeling with additive models selected papers ( updated 6/19/06) listening test ( temporary)

Last update: 7/22/2005

[back to members]