| Presentation | 2010-01-21 Online speaker clustering using an ergodic HMM and its application to meeting minute generation Takafumi KOSHINAKA, Kentaro NAGATOMO, Kenji SATOH, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | A novel online speaker clustering method suitable for real-time applications is proposed. Using an ergodic hidden Markov model (HMM), it employs incremental learning based on a variational Bayesian framework and provides probabilistic (non-deterministic) decisions for each input utterance, directly considering the specific history of preceding utterances. It makes possible more robust cluster estimation and precise classification of utterances than do conventional online methods. Experiments on meeting-speech data show that the proposed method produces 60-80% fewer classification errors than a conventional method does. They also show that it reduces speech recognition errors when combined with unsupervised speaker adaptation. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Hidden Markov models / variational Bayesian inference / model selection / meeting speech recognition |
| Paper # | CQ2009-62,PRMU2009-161,SP2009-102,MVE2009-84 |
| Date of Issue |
| Conference Information | |
| Committee | CQ |
|---|---|
| Conference Date | 2010/1/14(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 | Communication Quality (CQ) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Online speaker clustering using an ergodic HMM and its application to meeting minute generation |
| Sub Title (in English) | |
| Keyword(1) | Hidden Markov models |
| Keyword(2) | variational Bayesian inference |
| Keyword(3) | model selection |
| Keyword(4) | meeting speech recognition |
| 1st Author's Name | Takafumi KOSHINAKA |
| 1st Author's Affiliation | Common Platform Software Res. Labs., NEC Corporation() |
| 2nd Author's Name | Kentaro NAGATOMO |
| 2nd Author's Affiliation | Common Platform Software Res. Labs., NEC Corporation |
| 3rd Author's Name | Kenji SATOH |
| 3rd Author's Affiliation | Common Platform Software Res. Labs., NEC Corporation |
| Date | 2010-01-21 |
| Paper # | CQ2009-62,PRMU2009-161,SP2009-102,MVE2009-84 |
| Volume (vol) | vol.109 |
| Number (no) | 373 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |