| Presentation | 2010-05-26 Online Speech Separation based on Spectral Subtraction for Meeting Speech Recognition Yu NASU, Koichi SHINODA, Sadaoki FURUI, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | This paper proposes a speech separation method for meeting speech recognition, which operates in real time. The proposed method suppresses interfering components in speech signals using spectral subtraction, where the signals are recorded with tie-clip microphones attached to the lapels of the speakers. Compared with the other conventional methods, it requires less computational costs and the separated speech signals have less distortion. In a recognition task of simulated partly-overlapped speech, the proposed method improved the word accuracy from 66.5 % to 77.7 %. We also evaluated the method on speech data recorded in real meeting, and confirmed the improvement in accuracies of voice activity detection and speech recognition. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Sound source separation / Spectral subtraction / Voice activity detection / Meeting speech recognition |
| Paper # | EA2010-2,SIP2010-2,SP2010-2 |
| Date of Issue |
| Conference Information | |
| Committee | SIP |
|---|---|
| Conference Date | 2010/5/19(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 | Signal Processing (SIP) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Online Speech Separation based on Spectral Subtraction for Meeting Speech Recognition |
| Sub Title (in English) | |
| Keyword(1) | Sound source separation |
| Keyword(2) | Spectral subtraction |
| Keyword(3) | Voice activity detection |
| Keyword(4) | Meeting speech recognition |
| 1st Author's Name | Yu NASU |
| 1st Author's Affiliation | Tokyo Institute of Technology() |
| 2nd Author's Name | Koichi SHINODA |
| 2nd Author's Affiliation | Tokyo Institute of Technology |
| 3rd Author's Name | Sadaoki FURUI |
| 3rd Author's Affiliation | Tokyo Institute of Technology |
| Date | 2010-05-26 |
| Paper # | EA2010-2,SIP2010-2,SP2010-2 |
| Volume (vol) | vol.110 |
| Number (no) | 55 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |