Name
Abstract | ||
|---|---|---|
Because the norm of testing cepstral vector was shrinked in a noisy environment, the model parameters, i.e., mean vector and covariance matrix, should be adapted simultaneously. We propose a method called variance adapted likelihood measure (VALM) which adapts the mean vector using a projection-based scale factor and adapts the covariance matrix using a variance reduction function estimated from the training database. The variance reduction function can be obtained according to various phonetic units. In the hidden Markov model based experiments, the speech recognition performance is greatly improved by applying VALM. The most significant improvement is achieved when the variance reduction function is separately estimated for different state parameters. |
Year | DOI | Venue |
|---|---|---|
1996 | 10.1109/ICASSP.1996.540286 | ICASSP |
Keywords | Field | DocType |
different state parameter,noisy environment,likelihood measure,model parameter,hidden markov model,cepstral vector,noisy speech recognition,variance reduction function,covariance matrix,variance reduction,mean vector,databases,speech processing,hidden markov models,speech recognition,parameter estimation,adaptive signal processing,noise,testing | Speech processing,Estimation of covariance matrices,Pattern recognition,Markov model,Computer science,Speech recognition,Artificial intelligence,Adaptive filter,Covariance matrix,Estimation theory,Hidden Markov model,Variance reduction | Conference |
ISBN | Citations | PageRank |
0-7803-3192-3 | 2 | 0.46 |
References | Authors | |
3 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Jen-Tzung Chien | 1 | 918 | 82.45 |
| Lee-Min Lee | 2 | 46 | 8.10 |
| Hsiao-Chuan Wang | 3 | 370 | 64.93 |