Name
Abstract | ||
|---|---|---|
The projection-based likelihood measure, an e€ective means of reducing noise contamination in speech recognition, dynamically searches an optimal equalization factor for adapting the cepstral mean vector of hidden Markov model (HMM) to equalize the noisy observation. In this paper, we present a novel likelihood measure which extends the ad- aptation mechanism to the shrinkage of covariance matrix and the adaptation bias of mean vector. A set of adaptation functions is proposed for obtaining the compensation factors. Experiments indicate that the likelihood measure pro- posed herein can markedly elevate the recognition accuracy. Ó 1998 Elsevier Science B.V. All rights reserved. Zusammenfassung Das projektions-basierte Likelihood-Mass, das das Rauschen in der Spracherkennung e€ektiv verringert, sucht dynamisch einen optimalen Ausgleichfaktor, um den durchschnittlichen Cepstral-Vector des HMM an das kon- taminierte Signal auszupassen. In diesem Artikel prasentieren wir ein neues Likelihood-Mass, welches den An- passungsmechanismus auf das Schrumpfen der Kovarianzmatrix und die Anpassung des mittleren Bias-Vektors erweitert. Einige Adaptierungsfunktionen zur Berechnung der Kompensationsfaktoren werden vorgeschlagen. Experi- mente zeigen, daû das hier vorgeschlagene Likelihood-Mass die Erkennungsgenauigkeit wesentlich verbessern kann. Ó 1998 Elsevier Science B.V. All rights reserved. ReÂsume |
Year | DOI | Venue |
|---|---|---|
1998 | 10.1016/S0167-6393(98)00024-7 | Speech Communication |
Keywords | Field | DocType |
likelihood measure,noise interference,speech recognition,novel projection-based likelihood measure,hidden markov model,noisy speech recognition,robustness,covariance matrix | Colors of noise,Pattern recognition,Computer science,Markov model,Cepstrum,Signal-to-noise ratio,Speech recognition,Optimal estimation,White noise,Artificial intelligence,Covariance matrix,Hidden Markov model | Journal |
Volume | Issue | ISSN |
24 | 4 | Speech Communication |
Citations | PageRank | References |
1 | 0.42 | 11 |
Authors | ||
3 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Jen-Tzung Chien | 1 | 918 | 82.45 |
| Hsiao-Chuan Wang | 2 | 370 | 64.93 |
| Lee-Min Lee | 3 | 46 | 8.10 |