Presentation | 1999/7/8 Word recognition using a filter for extraction of power spectrum derivative Kinya Matsumoto, Tetsuo Funada, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | In noisy speech recognition using LPC cepstrum, recognition rate becomes very low. Therefore, in order to extract robust features of noisy speech, we use a filter designed for extraction of power spectrum derivative, which we call "Power Spectrum Derivative (PSD) method". In this study, we apply the proposed features to word recognition with continuous type HMMs, and compare the robustness of the features with LPC cepstrum, and BPFP features with we proposed in previous paper. In recognition experiments of noisy speech, PSD features show higher recognition rates than LPC cepstrum, and preform as well as BPFP one. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | spectrum derivative / FTTSS / threshold operation / robustness |
Paper # | SP99-42 |
Date of Issue |
Conference Information | |
Committee | SP |
---|---|
Conference Date | 1999/7/8(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) | Word recognition using a filter for extraction of power spectrum derivative |
Sub Title (in English) | |
Keyword(1) | spectrum derivative |
Keyword(2) | FTTSS |
Keyword(3) | threshold operation |
Keyword(4) | robustness |
1st Author's Name | Kinya Matsumoto |
1st Author's Affiliation | Graduate school of natural science and technology, kanazawa university() |
2nd Author's Name | Tetsuo Funada |
2nd Author's Affiliation | Faculty of engineering, kanazawa university |
Date | 1999/7/8 |
Paper # | SP99-42 |
Volume (vol) | vol.99 |
Number (no) | 165 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |