Gaussian mixture selection using context-independent HMM

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

Volumn 1, Issue , 2001, Pages 69-72

  Lee, Akinobu ^a   Kawahara, Tatsuya ^b   Shikano, Kiyohiro ^a  

^a NARA INSTITUTE OF SCIENCE AND TECHNOLOGY   (Japan)

^b KYOTO UNIVERSITY   (Japan)

Author keywords

[No Author keywords available]

Indexed keywords

ACOUSTIC LIKELIHOOD; CONTEXT INDEPENDENT MODELS; GAUSSIAN MIXTURES; GAUSSIAN SELECTION; HIDDEN MARKOV MODELS; PHONETIC TIED MIXTURE; TRIPHONE MODELS;

CALCULATIONS; COMPUTATIONAL COMPLEXITY; COMPUTATIONAL LINGUISTICS; MARKOV PROCESSES; MATHEMATICAL MODELS; PROBABILITY DISTRIBUTIONS; VECTOR QUANTIZATION;

CONTINUOUS SPEECH RECOGNITION;

EID: 0034857835     PISSN: 15206149     EISSN: None     Source Type: Journal    
DOI: 10.1109/ICASSP.2001.940769     Document Type: Article

References (7)

1. Vector quantization for efficient computation of continuous density likelihoods
2. Use of Gaussian selection in large vocabulary continuous speech recognition using HMM's
3. State-based Gaussian selection in large vocabulary continuous speech recognition using HMM's
4. An investigation of Gaussian shortlists
5. An efficient two-pass search algorithm using word trellis index
6. A new phonetic tied-mixture model for efficient decoding
7. Free software toolkit for Japanese large large vocabulary continuous speech recognition