DocumentCode :
302341
Title :
Two-pass strategy for continuous speech recognition with detection and transcription of unknown words
Author :
Matsunaga, Shoichi ; Sakamoto, Hiroyuki
Author_Institution :
ATR Interpreting Telecommun. Res. Labs., Kyoto, Japan
Volume :
1
fYear :
1996
fDate :
7-10 May 1996
Firstpage :
538
Abstract :
This paper proposes a new approach of using a two-pass strategy to effectively recognize continuous speech including unknown words. In this approach, the first pass uses context-independent phoneme HMMs to recognize registered words and phoneme-cluster HMMs to detect unknown words. In the second pass context-dependent phone models are used for precise recognition where unknown words are transcribed. In sentence recognition experiments using this unknown-word processing, phoneme cluster models that consider the Japanese syllabic construction achieved a higher word accuracy rate of 70.3%, compared with 59.2% for sentence recognition without this processing. Furthermore, the amount of processing was reduced by about half compared with a detection method using phoneme HMMs. The total system achieves a 75.2% phoneme accuracy rate including the transcription of unknown words
Keywords :
context-sensitive grammars; hidden Markov models; natural languages; speech processing; speech recognition; Japanese syllabic construction; context dependent phone models; context independent phoneme HMM; continuous speech recognition; phoneme accuracy rate; phoneme cluster HMM; phoneme cluster models; sentence recognition experiments; two-pass strategy; unknown word processing; unknown words detection; unknown words transcription; word accuracy rate; Acoustic signal detection; Context modeling; Error analysis; Hidden Markov models; Speech processing; Speech recognition; Text processing;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1996. ICASSP-96. Conference Proceedings., 1996 IEEE International Conference on
Conference_Location :
Atlanta, GA
ISSN :
1520-6149
Print_ISBN :
0-7803-3192-3
Type :
conf
DOI :
10.1109/ICASSP.1996.541152
Filename :
541152
Link To Document :
بازگشت