| Presentation | 2010/1/14 Multimodal speech recognition using multimodal voice activity detection Satoshi TAMURA, Masato ISHIKAWA, Takashi HASHIBA, Shin'ichi TAKEUCHI, Satoru HAYAMIZU, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Audio-Visual Automatic Speech Recognition (AVASR) has been developed to enhance the robustness in noisy environments, using visual information in addition to acoustic features. Similarly, Audio-Visual Voice Activity Detection (AVVAD) has been investigated and used to increase the precision of VAD, since detecting presence of speech in noisy audio signals contributes ASR performance. In this paper, we propose a novel speech recognition method combining AVASR and AVVAD: combinations of model-based and model-free, and feature-fusion-based or decision-fusion-based methods. To evaluate the proposed schemes, recognition experiments were conducted using noisy audio-visual data. Then it is found that the proposed method using the model-free feature-fusion AVVAD method outperforms not only audio-only ASR but also conventional AVASR. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | multimodal / speech recognition / voice activity detection / feature fusion / decision fusion |
| Paper # | CQ2009-105,PRMU2009-204,SP2000-145,MVE2009-127 |
| 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) | Multimodal speech recognition using multimodal voice activity detection |
| Sub Title (in English) | |
| Keyword(1) | multimodal |
| Keyword(2) | speech recognition |
| Keyword(3) | voice activity detection |
| Keyword(4) | feature fusion |
| Keyword(5) | decision fusion |
| 1st Author's Name | Satoshi TAMURA |
| 1st Author's Affiliation | Faculty of Engineering, Gifu University() |
| 2nd Author's Name | Masato ISHIKAWA |
| 2nd Author's Affiliation | Graduated School of Engineering, Gifu University |
| 3rd Author's Name | Takashi HASHIBA |
| 3rd Author's Affiliation | Graduated School of Engineering, Gifu University |
| 4th Author's Name | Shin'ichi TAKEUCHI |
| 4th Author's Affiliation | Virtual System Laboratory, Gifu University |
| 5th Author's Name | Satoru HAYAMIZU |
| 5th Author's Affiliation | Faculty of Engineering, Gifu University |
| Date | 2010/1/14 |
| Paper # | CQ2009-105,PRMU2009-204,SP2000-145,MVE2009-127 |
| Volume (vol) | vol.109 |
| Number (no) | 373 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |