Presentation | 2006/12/15 Using presentation slide information for lecture speech recognition Hiroki YAMAZAKI, Koji IWANO, Koichi SHINODA, Sadaoki FURUI, Haruo YOKOTA, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | We propose a dynamic language model adaptation method for lecture speech recognition in which the information of text on slides for lectures is used. The speech data corresponding to each slide are recognized with a language model adapted to them by using the slide texts as adaptation data. We evaluated the proposed method by using the speech data of three classroom courses in Japanese, and confirmed its effectiveness. The average speech recognition error was reduced by 3.1% by the global adaptation using all slides used in a cource. The error rates of recall and precision for keywords were also reduced by 21.5% and 13.8% respectively. Furthermore, we achieved the improvement of keyword detection performance by the adaptation using each slide locally. The error rates of recall and precision for keywords were reduced by 3.1% and 1.4% respectively from global adaptation. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Language model adaptation / speech recognition / classroom lecture speech |
Paper # | NLC2006-66,SP2006-122 |
Date of Issue |
Conference Information | |
Committee | NLC |
---|---|
Conference Date | 2006/12/15(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) | Using presentation slide information for lecture speech recognition |
Sub Title (in English) | |
Keyword(1) | Language model adaptation |
Keyword(2) | speech recognition |
Keyword(3) | classroom lecture speech |
1st Author's Name | Hiroki YAMAZAKI |
1st Author's Affiliation | Department of Computer Science, Tokyo Institute of Technology() |
2nd Author's Name | Koji IWANO |
2nd Author's Affiliation | Department of Computer Science, Tokyo Institute of Technology |
3rd Author's Name | Koichi SHINODA |
3rd Author's Affiliation | Department of Computer Science, Tokyo Institute of Technology |
4th Author's Name | Sadaoki FURUI |
4th Author's Affiliation | Department of Computer Science, Tokyo Institute of Technology |
5th Author's Name | Haruo YOKOTA |
5th Author's Affiliation | Department of Computer Science, Tokyo Institute of Technology |
Date | 2006/12/15 |
Paper # | NLC2006-66,SP2006-122 |
Volume (vol) | vol.106 |
Number (no) | 442 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |