Multiple-regression hidden markov model

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

Volumn 1, Issue , 2001, Pages 513-516

  Fujinaga, K ^a   Nakai, M ^a   Shimodaira, H ^a   Sagayama, S ^a  

^a JAPAN ADVANCED INSTITUTE OF SCIENCE AND TECHNOLOGY   (Japan)

Author keywords

[No Author keywords available]

Indexed keywords

ACOUSTIC FEATURES; HIDDEN MARKOV MODEL; PHONEMES;

MARKOV PROCESSES; PROBABILITY DISTRIBUTIONS; REGRESSION ANALYSIS; SPEECH ANALYSIS; VECTORS;

SPEECH RECOGNITION;

EID: 0034855363     PISSN: 15206149     EISSN: None     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (10)

1. A study on speaker adaptation of the parameters of continuous density hidden Markov models
2. Speaker adaptation based on transfer vector field smoothing with continuous mixture density HMMs
3. Speaker adaptation of HMM's using linear regression
4. A successive state splitting algorithm for efficient allophone modeling
5. On the influence of hyperarticulated speech on the recognition performance
6. Specialized acoustic models for hyperarticulated speech
7. Pitch dependent phone modeling for HMM based speech recognition
8. Mixture autoregressive hidden Markov models for speech signals
9. Relationship between pitch control and vowel articulation
10. HMM topology design using maximum likelihood successive state splitting