Our research interests include spoken language processing, natural language processing, multimedia information retrieval, and pattern recognition.
Communicating with computers using speech has been a dream of many people since the invention of computers. Progress
towards realizing this dream has been slow, but steady, through the development of systems supporting voice commands, dictation,
text-to-speech synthesis, and human-computer spoken dialogue. Speech recognition, speech synthesis, language understanding,
dialogue management, etc. are crucial to the development of human-computer speech interface. Our research has
focused mainly on speech recognition, speech synthesis, and speaker recognition.
Huge collections of video and audio recordings, which have captured events of the last century, remain an untapped resource
with historical and scientific value. Moreover, multimedia technology and the Internet are creating a completely new information
era in which an exponentially increasing number of audio documents are being accumulated and made available. Accordingly,
there are many digital library projects worldwide on how multimedia digital libraries can be established and used. We have been
studying spoken document processing, indexing and the retrieval of Mandarin broadcast news for several years and have developed
several basic technologies, as well as prototype retrieval systems. More recently, we have extended our studies to content-based music information
retrieval. Our research has focused mainly on query by singing/humming, solo vocal modeling, and music tag annotation and retrieval. Our future plans
include further improvement of the speech and music information retrieval technology.