| Presentation | 2014/12/8 Consideration on Age- and Gender-independent Speech Recognition using DNN-HMM HIROSHI SEKI, KAZUMASA YAMAMOTO, SEIICHI NAKAGAWA, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | We have studied syllable-based acoustic modeling for Japanese speech recognition. In this paper, we first investigate the performance of recognition accuracy using phoneme/syllable-based DNN-HMM. The results show that there's no significant difference between phoneme/syllable-based DNN-HMM. Second, we investigate the age- and gender-independent speech recognition using DNN-HMM. We use three types of corpora(adult, elder, child), and each corpus contains male and female speech data. In general, speaker-independent system cannot handle the specific information of speakers, and the recognition performance of speaker independent model is lower that of speaker dependent model. Our experimental results show that one DNN-HMM trained by all corpora with a class-dependent feature normalization method achieves better performance compared to class-dependent DNN-HMMs. Finally, we investigate the incorporation of information on corpora into DNN. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Deep Neural Network / HMM / DNN-HMM / speaker independent speech recognition |
| Paper # | Vol.2014-SLP-104 No.29 |
| Date of Issue |
| Conference Information | |
| Committee | SP |
|---|---|
| Conference Date | 2014/12/8(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) | Consideration on Age- and Gender-independent Speech Recognition using DNN-HMM |
| Sub Title (in English) | |
| Keyword(1) | Deep Neural Network |
| Keyword(2) | HMM |
| Keyword(3) | DNN-HMM |
| Keyword(4) | speaker independent speech recognition |
| 1st Author's Name | HIROSHI SEKI |
| 1st Author's Affiliation | Toyohashi University of Technology() |
| 2nd Author's Name | KAZUMASA YAMAMOTO |
| 2nd Author's Affiliation | Toyohashi University of Technology |
| 3rd Author's Name | SEIICHI NAKAGAWA |
| 3rd Author's Affiliation | Toyohashi University of Technology |
| Date | 2014/12/8 |
| Paper # | Vol.2014-SLP-104 No.29 |
| Volume (vol) | vol.114 |
| Number (no) | 365 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |