Upper and lower bounds on the mean of noisy speech: Application to minimax classification

IEEE Transactions on Speech and Audio Processing

Volumn 10, Issue 2, 2002, Pages 79-88

  Afify, Mohamed ^b   Siohan, Olivier ^a,b   Lee, Chin Hui ^a,b  

^a IEEE

^b LUCENT TECHNOLOGIES   (United States)

Author keywords

Minimax classification; Robust decision rules; Speech recognition

Indexed keywords

CLASSIFICATION (OF INFORMATION); DECISION THEORY; SIGNAL TO NOISE RATIO; SPEECH; SPEECH ANALYSIS; SPEECH PROCESSING; SPEECH SYNTHESIS; SPURIOUS SIGNAL NOISE;

ADDITIVE NOISE; MEL FREQUENCY CEPSTRAL COEFFICIENT; MINIMAX AND BAYESIAN PREDICTIVE CLASSIFICATION METHODS; MINIMAX CLASSIFICATION; NOISY SPEECH; PARAMETRIC NEIGHBORHOOD; PLUG-IN MAXIMUM A POSTERIORI; ROBUST DECISION RULES; SIGNAL ENHANCEMENT; VECTOR TAYLOR SERIES;

SPEECH RECOGNITION;

EID: 0036475998     PISSN: 10636676     EISSN: None     Source Type: Journal    
DOI: 10.1109/89.985545     Document Type: Article

References (14)

1. Speech recognition in noisy environments: A survey
2. A vector Taylor series approach for environment-independent speech recognition
3. Statistical linear approximation for environment compensation
4. Speech recognition using noise-adaptive prototypes
5. Robust speech recognition based on adaptive classification and decision strategies
6. On adaptive decision rules and decision parameter adaptation for automatic speech recognition
7. Robust speech recognition based on a Bayesian prediction approach
8. A minimax classification approach with application to robust speech recognition
9. A minimax search algorithm for robust continuous speech recognition
10. Comparison of parametric representation for monosyllabic word recognition in continuously spoken sentences
11. Sequential noise estimation with optimal forgetting for robust speech recognition
12. Cepstral parameter compensation for HMM recognition in noise