4th International Conference on Spoken Language Processing

Philadelphia, PA, USA
October 3-6, 1996

Modeling Context-dependent Phonetic Units in a Continuous Speech Recognition System for Mandarin Chinese

Jim Jian-Xiong Wu (1,2), Li Deng (2), Jacky Chan (2)

We study the problem of phonetic modeling for continuous Mandarin speech recognition by providing a systematic performance comparison for systems based on following primitive speech units: syllable, demi-syllable (Initials and Finals), context-independent phones, left-or-right context-dependent phones (diphones), and left-and-right context-dependent phones (triphones). In our speaker-dependent continuous speech recognition experiments, a generalized triphone system has achieved the best performance among all. Our best system contrasts most other Mandarin speech recognition systems which have been based on demi-syllable units.

Full Paper

Bibliographic reference.  Wu, Jim Jian-Xiong / Deng, Li / Chan, Jacky (1996): "Modeling context-dependent phonetic units in a continuous speech recognition system for Mandarin Chinese", In ICSLP-1996, 2281-2284.