Name
Title | ||
|---|---|---|
Minimum generation error criterion considering global/local variance for HMM-based speech synthesis |
Abstract | ||
|---|---|---|
Due to the inconsistency between the maximum likelihood (ML) based training and the synthesis application in HMM-based speech synthesis, a minimum generation error (MGE) criterion had been proposed for HMM training. This paper continues to apply the MGE criterion to model adaptation for HMM-based speech synthesis. We propose a MGE linear regression (MGELR) based model adaptation algorithm, where the regression matrices used to transform source models to target models are optimized to minimize the generation errors for the input speech data uttered by the target speaker. The proposed MGELR approach was compared with the maximum likelihood linear regression (MLLR) based model adaptation. Experimental results indicate that the generation errors were reduced after the MGELR-based model adaptation. And from the subjective listening test, the discrimination and the quality of the synthesized speech using MGELR were better than the results using MLLR. |
Year | DOI | Venue |
|---|---|---|
2008 | 10.1109/ICASSP.2008.4518686 | ICASSP |
Keywords | Field | DocType |
maximum likelihood based training,speech synthesis,minimum generation error,minimum generation error criterion,local variance,regression analysis,maximum likelihood estimation,error analysis,model adaptation,transform source models,hmm,global variance,maximum likelihood linear regression,minimum generation error linear regression,model adaptation algorithm,generation error minimization,hidden markov models,hmm-based speech synthesis,maximum likelihood,generalization error,linear regression | Speech synthesis,Pattern recognition,Regression,Matrix (mathematics),Regression analysis,Computer science,Local variance,Maximum likelihood,Speech recognition,Artificial intelligence,Hidden Markov model,Linear regression | Conference |
ISSN | ISBN | Citations |
1520-6149 E-ISBN : 978-1-4244-1484-0 | 978-1-4244-1484-0 | 12 |
PageRank | References | Authors |
0.86 | 7 | 5 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Long Qin | 1 | 27 | 3.93 |
| Yi-Jian Wu | 2 | 288 | 27.12 |
| Zhen-Hua Ling | 3 | 850 | 83.08 |
| Ren-Hua Wang | 4 | 344 | 41.36 |
| Li-Rong Dai | 5 | 1070 | 117.92 |