SUBSPACE DISTRIBUTION CLUSTERING FOR CONTINUOUS OBSERVATION DENSITY HIDDEN MARKOV MODELS

5th European Conference on Speech Communication and Technology, EUROSPEECH 1997

Volumn , Issue , 1997, Pages 107-110

  Bocchieri, Enrico ^a   Mak, Brian ^b  

^a AT AND T LABS RESEARCH   (United States)

^b OREGON HEALTH AND SCIENCE UNIVERSITY   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

GAUSSIAN DISTRIBUTION; PROBABILITY DENSITY FUNCTION; SPEECH COMMUNICATION; TRELLIS CODES;

B CLUSTERING; CLUSTERINGS; COMPUTATIONAL COSTS; CONTINUOUS OBSERVATION; FEATURE SPACE; GAUSSIAN-MIXTURES; HIDDEN-MARKOV MODELS; HIGH COMPUTATION EFFICIENCY; HIGH MEMORY EFFICIENCIES; STATE PROBABILITY;

HIDDEN MARKOV MODELS;

EID: 85135158363     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (5)

1. S.Takahashi, S.Sagayama, "Four-Level Tied-Structure for Efficient Representation of Acoustic Modeling," Proc. ICASSP-95, pp. 520-523.
2. S.Takahashi, S.Sagayama, "Effects of Variance Tying for Four-Level Tied Structure Phone Models," Proc ASI, March 95, pp 141-142.
3. E.Bocchieri, G.Riccardi "State tying of triphone HMM's for the 1994 AT&T ARPA ATIS Recognizer," Proc Eurospeech 95, pp 1499-1502.
4. E.Bocchieri, "Vector Quantization For Efficient Computation Of Continuous Density Likelihoods", Proc. ICASSP-93, vol II, pp 692-695.
5. K.Knill, M Gales, S.Young, "Use of Gaussian Selection In Large Vocabulary Continuous Speech Recognition Using HMM'S", Proc. ICSLP-96, vol 1, pp 470-473.