| Presentation | 2004/12/13 Variational Bayesian Based Topology Training and Mixture Component Splitting for Acoustic Modeling Takatoshi JITSUHIRO, Satoshi NAKAMURA, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | We propose a automatic generation method of non-uniform and context-dependent HMM topology and a splitting method of mixture components based on the Variational Bayesian (VB) approach. Although the Maximum' Likelihood (ML) criterion is generally used to create HMM topologies, it has an over-fitting problem. Recently, to avoid this problem, the VB approach has been applied to create acoustic models for speech recognition. We introduce the VB approach to the Successive State Splitting (SSS) algorithm, which can create both contextual and temporal variations for HMMs. Experimental results show that the proposed method can automatically create more efficient models than those by the original method. We employ the VB approach to increase the number of mixture components. The VB approach obtained almost the same performance with the smaller number of mixture components in comparison with that obtained by using ML-based methods. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | speech recognition / acoustic model / topology training / SSS algorithm / Variational Bayesian approach |
| Paper # | NLC2004-51,SP2004-91 |
| Date of Issue |
| Conference Information | |
| Committee | NLC |
|---|---|
| Conference Date | 2004/12/13(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Natural Language Understanding and Models of Communication (NLC) |
|---|---|
| Language | ENG |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Variational Bayesian Based Topology Training and Mixture Component Splitting for Acoustic Modeling |
| Sub Title (in English) | |
| Keyword(1) | speech recognition |
| Keyword(2) | acoustic model |
| Keyword(3) | topology training |
| Keyword(4) | SSS algorithm |
| Keyword(5) | Variational Bayesian approach |
| 1st Author's Name | Takatoshi JITSUHIRO |
| 1st Author's Affiliation | ATR Spoken Language Translation Research Laboratories() |
| 2nd Author's Name | Satoshi NAKAMURA |
| 2nd Author's Affiliation | ATR Spoken Language Translation Research Laboratories |
| Date | 2004/12/13 |
| Paper # | NLC2004-51,SP2004-91 |
| Volume (vol) | vol.104 |
| Number (no) | 538 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |