| Presentation | 2015-03-02 Two-Channel Audio Source Separation in Reverberant Conditions Based on Single Voice Activity Segments Atsushi MATSUDA, Arata KAWAMURA, Youji IIGUNI, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | This paper presents a two-channel audio source separation method in reverberant conditions based on single voice activity segments. Since the conventional method uses all segments of observed signals, there exist a lot of noise in the observed signals. The noise make it difficult to perform clustering for the observed signals. Therefore, the conventional method needs a prior knowledge about the number of mixing sources, and the entire process is computationally heavy. On the other hand, the proposed method uses single voice activity segments in order to reduce the noise. This makes it possible to reduce computational quantity of a two-channel audio source separation in reverberant conditions, even if the number of sources is unknown. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Two-channel audio source separation in reverberant conditions / Single voice activity segments / Number of sources / Reduction in computational quantity |
| Paper # | EA2014-81,SIP2014-122,SP2014-144 |
| Date of Issue |
| Conference Information | |
| Committee | SIP |
|---|---|
| Conference Date | 2015/2/23(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) | Two-Channel Audio Source Separation in Reverberant Conditions Based on Single Voice Activity Segments |
| Sub Title (in English) | |
| Keyword(1) | Two-channel audio source separation in reverberant conditions |
| Keyword(2) | Single voice activity segments |
| Keyword(3) | Number of sources |
| Keyword(4) | Reduction in computational quantity |
| 1st Author's Name | Atsushi MATSUDA |
| 1st Author's Affiliation | Graduate School of Engineering Science, Osaka University() |
| 2nd Author's Name | Arata KAWAMURA |
| 2nd Author's Affiliation | Graduate School of Engineering Science, Osaka University |
| 3rd Author's Name | Youji IIGUNI |
| 3rd Author's Affiliation | Graduate School of Engineering Science, Osaka University |
| Date | 2015-03-02 |
| Paper # | EA2014-81,SIP2014-122,SP2014-144 |
| Volume (vol) | vol.114 |
| Number (no) | 474 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |