| Presentation | 2004/12/14 Speech Recognition with Integration of Confidence Measures Akio KOBAYASHI, Kazuo ONOE, Shoe SATO, Toru IMAI, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | This paper describes speech recognition using integration of confidence measures. Confidence measures are widely used in classifying word hypotehses into correct or incorrect classes. The measures will make a large contribution to the improvement of the word error rates (WERs) when they give high classification performances. In this paper we design an integrated confidence measure by maximum entropy (ME) modeling for the classifier. We propose an n-best rescoring method using output scores of the classifier to improve the WERs of the. system by correcting the acoustic and language scores. Our proposed method was compared with the classifier based on support vector machines (SVMs), which also integrates the confidence measures. The results of classification and n-best rescoring showed the ME-based classifier performed better than the SVM-based classifier. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | words confidence measure / maximum entropy models / support vector machines / n-best rescoring |
| Paper # | NLC2004-72,SP2004-112 |
| Date of Issue |
| Conference Information | |
| Committee | NLC |
|---|---|
| Conference Date | 2004/12/14(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 | Natural Language Understanding and Models of Communication (NLC) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Speech Recognition with Integration of Confidence Measures |
| Sub Title (in English) | |
| Keyword(1) | words confidence measure |
| Keyword(2) | maximum entropy models |
| Keyword(3) | support vector machines |
| Keyword(4) | n-best rescoring |
| 1st Author's Name | Akio KOBAYASHI |
| 1st Author's Affiliation | NHK Science & Technical Research Laboratories() |
| 2nd Author's Name | Kazuo ONOE |
| 2nd Author's Affiliation | NHK Science & Technical Research Laboratories |
| 3rd Author's Name | Shoe SATO |
| 3rd Author's Affiliation | NHK Science & Technical Research Laboratories |
| 4th Author's Name | Toru IMAI |
| 4th Author's Affiliation | NHK Science & Technical Research Laboratories |
| Date | 2004/12/14 |
| Paper # | NLC2004-72,SP2004-112 |
| Volume (vol) | vol.104 |
| Number (no) | 539 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |