Compact acoustic model for embedded implementation

8th International Conference on Spoken Language Processing, ICSLP 2004

Volumn , Issue , 2004, Pages 693-696

  Park, Junho ^a   Ko, Hanseok ^a  

^a Korea University   (South Korea)

Author keywords

[No Author keywords available]

Indexed keywords

DESIRABLE FEATURES; EMBEDDED IMPLEMENTATION; EVALUATION RESULTS; MEMORY PROBLEMS; NOCV2; PERFORMANCE DEGRADATION; QUANTIZING METHOD; SPEECH RECOGNITION SYSTEMS; SYSTEM RESOURCES;

SPEECH RECOGNITION;

EID: 85009080861     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (6)

1. Young, S. J., "The general use of tying in phoneme-based HMM speech recognizers," ICASSP1992, pp. 569-572.
2. Lee, A., et al, "A new phonetic tied-mixture model for efficient decoding," ICASSP2000, pp. 1269-1272.
3. Bocchieri, E. and Mak, B. K-W., "Subspace distribution clustering hidden Markov model," IEEE Trans on speech and Audio Processing, vol. 9, March 2001, pp. 264-276.
4. Duchateau, J., et al, "Fast and accurate acoustic modeling with semi-continuous HMMs," Speech Communication, vol 24, 1998, pp. 5-17.
5. Gales, M., et al, "State-based Gaussian selection in large vocabulary continuous speech recognition using HMM's," IEEE Trans on speech and Audio Processing, vol. 7, March 1999, pp.152-161.
6. Kim, K.H., et al, "Usage of exclusive HMMs for Korean-Digit in the embedded ASR system using tied-mixture method," Submitted to Proc. ICSLP'04, 2004.