| Presentation | 2002/12/12 Acoustic and Linguistic Modeling for Lecture Speech Recognition Ryousuke TSUTSUIMI, Masaharu KATOH, Tetsuo KOSAKA, Masaki KOHDA, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Large vocabulary continuous speech recognition (LVCSR) became practical application level for a newspaper read-speech. However, in LVCSR of spontaneous utterance such as a lecture speech etc. there exist many problems which make recognition difficult. The purpose of this paper is to make precise acoustic and linguistic models for LVCSR system of spontaneous utterance. Speech training data selection is performed to make acoustic model, and morpheme analysis of pronunciation variant dependency is executed to make linguistic model. WER of 31.4% was obtained without speaker adaptation of acoustic model. Trigram and 4-gram with various cutoffs were introduced to rescore language likelifood. WER inprovement by introducing cutoffs was 0.6%. Furthermore, speaker adaptation of acoustic model was carried out, and WER improvement of 3.2% was obtained. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | lecture speech / acoustic model / speech training data selection / linguistic model / pronunciation variant dependency / speaker adaptation. |
| Paper # | SP2002-140 |
| Date of Issue |
| Conference Information | |
| Committee | SP |
|---|---|
| Conference Date | 2002/12/12(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 | Speech (SP) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Acoustic and Linguistic Modeling for Lecture Speech Recognition |
| Sub Title (in English) | |
| Keyword(1) | lecture speech |
| Keyword(2) | acoustic model |
| Keyword(3) | speech training data selection |
| Keyword(4) | linguistic model |
| Keyword(5) | pronunciation variant dependency |
| Keyword(6) | speaker adaptation. |
| 1st Author's Name | Ryousuke TSUTSUIMI |
| 1st Author's Affiliation | Faculty of Engineering, Yamagata University() |
| 2nd Author's Name | Masaharu KATOH |
| 2nd Author's Affiliation | Faculty of Engineering, Yamagata University |
| 3rd Author's Name | Tetsuo KOSAKA |
| 3rd Author's Affiliation | Faculty of Engineering, Yamagata University |
| 4th Author's Name | Masaki KOHDA |
| 4th Author's Affiliation | Faculty of Engineering, Yamagata University |
| Date | 2002/12/12 |
| Paper # | SP2002-140 |
| Volume (vol) | vol.102 |
| Number (no) | 529 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |