Abstract | ||
---|---|---|
This paper proposes a state duration modeling method using full covariance matrix for HMM-based speech synthesis. In this method, a full covariance matrix instead of the conventional diagonal covariance matrix is adopted in the multi-dimensional Gaussian distribution to model the state duration of each context-dependent phoneme. At synthesis stage, the state durations are predicted using the clustered context-dependent distributions with full covariance matrices. Experimental results show that the synthesized speech using full-covariance state duration models is more natural than the conventional method when we change the speaking rate of synthesized speech. |
Year | DOI | Venue |
---|---|---|
2009 | 10.1109/ICASSP.2009.4960513 | ICASSP |
Keywords | Field | DocType |
hmm-based speech synthesis,state duration modeling method,conventional method,synthesized speech,full covariance state duration,state duration,full covariance matrix,full-covariance state duration model,conventional diagonal covariance matrix,context-dependent distribution,context-dependent phoneme,hidden markov models,computer science,gaussian distribution,duration,context modeling,covariance matrix,hmm,speech,hidden markov model,speech synthesis,flowcharts,high temperature superconductors,context dependent,mathematical model,predictive models,training data | Covariance function,Speech synthesis,Estimation of covariance matrices,Pattern recognition,Matrix (mathematics),Covariance intersection,Speech recognition,Artificial intelligence,Covariance matrix,Hidden Markov model,Mathematics,Covariance | Conference |
ISSN | Citations | PageRank |
1520-6149 | 2 | 0.49 |
References | Authors | |
5 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Heng Lu | 1 | 58 | 9.22 |
Yi-Jian Wu | 2 | 288 | 27.12 |
Keiichi Tokuda | 3 | 3016 | 250.00 |
Li-Rong Dai | 4 | 1070 | 117.92 |
Ren-Hua Wang | 5 | 344 | 41.36 |