Presentation | 2016-11-18 Speech enhancement in the specific area using information of sound source position Takuto Yoshimizu, Akitoshi Kataoka, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | As technology to emphasize the desired signal while to adapt to the noise environment, the beamformer using a microphone array has been proposed. The beamformer is technology to emphasize the desired signal by aligning the phase of the incoming signal from desired direction. In addition, we have confirm that it is possible to improve noise suppression amount by to use a Wiener post-filter together with a beamformer. However, these methods deteriorate to performance, because it is difficult to suppress incoming noise from the direction same as desired sound source in the environment such as the desired sound source is surrounded by noise. The methods which emphasize desired sound source using two microphone arrays arranged in different positions are proposed. We can get different information of sound by using two microphone arrays. They are methods that emphasize desired signal using that information. However, there is problem that desired signal is distorted due to the estimation error of the acoustic model. This paper, we propose the suppression method in Wiener post-filter based on the position information of the sound source. Acoustic model used in the Wiener post-filter is estimated from the matrix of beam patterns corresponding to the power spectrum and the sound source position of the beamformer output emphasized sound source position. Because the matrix of beam patterns is regular matrix, this method can be smaller the error of estimated acoustic model than conventional method. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | microphone array / noise reduction / beamformer / Wiener post-filter |
Paper # | EA2016-65,EMM2016-71 |
Date of Issue | 2016-11-10 (EA, EMM) |
Conference Information | |
Committee | EA / EMM |
---|---|
Conference Date | 2016/11/17(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Compal Hall (Oita) |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | Multimedia Fusion, Content Processing, Multimedia Retrieval, Digital Watermarking, and Related Topics |
Chair | Mitsunori Mizumachi(Kyushu Inst. of Tech.) / Akinori Ito(Tohoku Univ.) |
Vice Chair | Yoichi Haneda(Univ. of Electro-Comm.) / Suehiro Shimauchi(NTT) / Masaki Kawamura(Yamaguchi Univ.) / Hirohisa Hioki(Kyoto Univ.) |
Secretary | Yoichi Haneda(KDDI R&D Labs.) / Suehiro Shimauchi(Akita Pref. Univ.) / Masaki Kawamura(Nagasaki Univ.) / Hirohisa Hioki(Osaka Pref. Univ.) |
Assistant | Shigeto Takeoka(Shizuoka Inst. of Science and Tech.) / TREVINO Jorge(Tohoku Univ.) / Rui Shogenji(Shizuoka Univ.) / Masaaki Fujiyoshi(Tokyo Metropolitan Univ.) |
Paper Information | |
Registration To | Technical Committee on Engineering Acoustics / Technical Committee on Enriched MultiMedia |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Speech enhancement in the specific area using information of sound source position |
Sub Title (in English) | |
Keyword(1) | microphone array |
Keyword(2) | noise reduction |
Keyword(3) | beamformer |
Keyword(4) | Wiener post-filter |
Keyword(5) | |
1st Author's Name | Takuto Yoshimizu |
1st Author's Affiliation | Ryukoku University(Ryukoku Univ.) |
2nd Author's Name | Akitoshi Kataoka |
2nd Author's Affiliation | Ryukoku University(Ryukoku Univ.) |
Date | 2016-11-18 |
Paper # | EA2016-65,EMM2016-71 |
Volume (vol) | vol.116 |
Number (no) | EA-302,EMM-303 |
Page | pp.pp.101-106(EA), pp.101-106(EMM), |
#Pages | 6 |
Date of Issue | 2016-11-10 (EA, EMM) |