Paper Info
Title
Modeling Context-Dependent Phonetic Units In A Continuous Speech Recognition System For Mandarin Chinese
Abstract
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.
Year
DOI
Venue
1996
10.1109/ICSLP.1996.607262
ICSLP 96 - FOURTH INTERNATIONAL CONFERENCE ON SPOKEN LANGUAGE PROCESSING, PROCEEDINGS, VOLS 1-4
Keywords
Field
DocType
speech recognition,underwater acoustics,linguistics,training data,business,hidden markov models,context dependent,mandarin chinese,context modeling,acoustical engineering,natural languages
Triphone,Speech corpus,Speech synthesis,Computer science,Audio mining,Speech recognition,Speaker recognition,Natural language processing,Artificial intelligence,VoxForge,Speech technology,Mandarin Chinese
Conference
Citations 
PageRank 
References 
7
1.13
5
Authors
3
Name
Order
Citations
PageRank
Jim Jian-Xiong Wu1101.92
Li Deng2124.32
Jacky Chan391.51