A Bayesian approach to speech feature enhancement using the dynamic cepstral prior

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

Volumn 1, Issue , 2002, Pages

  Deng, Li ^a   Droppo, Jasha ^a   Acero, Alex ^a  

^a MICROSOFT RESEARCH   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

ACOUSTIC DISTORTION; ALGORITHMS; FEATURE EXTRACTION; LEAST SQUARES APPROXIMATIONS; PROBABILITY; STATISTICAL METHODS; VECTORS;

BAYESIAN ESTIMATION; DYNAMIC CEPSTRAL PRIOR; MINIMUM MEAN SQUARE ERROR;

SPEECH RECOGNITION;

EID: 0036299277     PISSN: 15206149     EISSN: None     Source Type: Conference Proceeding    
DOI: 10.1109/icassp.2002.5743867     Document Type: Conference Paper

References (14)

1. HMM adaptation using vector Taylor series for noisy speech recognition
2. Speech denoising and dereverberation using probabilistic models
3. A new method for speech denoising and robust speech recognition using probabilistic models for clean speech and for souse
4. Large-vocabulary speech recognition under adverse acoustic environments
5. High-performance robust speech recognition using stereo training data
6. Recursive estimation of nonstationary noise using a nonlinear model with iterative stochastic approximation
7. Evaluation of the SPLICE algorithm on the Aurora2 database
8. Statistical-model-based speech enhancement systems
9. ALGONQUIN: Iterating Laplace's method to remove multiple types of acoustic distortion for robust speech recognition
10. Nonstationary environment compensation based on sequential estimation
11. A vector Taylor series approach for environment-independent speech recognition
12. The AURORA experimental framework for the performance evaluations of speech recognition systems under noisy conditions
13. HMM-based strategies for enhancement of speech embedded in nonstationary noise
14. Model-based compensation of the additive noise for continuous speech recognition