Practically Efficient Blind Speech Separation Using Frequency Band Selection Based on Magnitude Squared Coherence and a Small Dodecahedral Microphone Array
Small agglomerative microphone array systems have been proposed for use with speech communication and recognition systems. Blind source separation methods based on frequency domain independent component analysis have shown significant separation performance, and the microphone arrays are small enoug...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2012-01-01
|
| Series: | Journal of Electrical and Computer Engineering |
| Online Access: | http://dx.doi.org/10.1155/2012/324398 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832554554740178944 |
|---|---|
| author | Kazunobu Kondo Yusuke Mizuno Takanori Nishino Kazuya Takeda |
| author_facet | Kazunobu Kondo Yusuke Mizuno Takanori Nishino Kazuya Takeda |
| author_sort | Kazunobu Kondo |
| collection | DOAJ |
| description | Small agglomerative microphone array systems have been proposed for use with speech communication and recognition systems. Blind source separation methods based on frequency domain independent component analysis have shown significant separation performance, and the microphone arrays are small enough to make them portable. However, the level of computational complexity involved is very high because the conventional signal collection and processing method uses 60 microphones. In this paper, we propose a band selection method based on magnitude squared coherence. Frequency bands are selected based on the spatial and geometric characteristics of the microphone array device which is strongly related to the dodecahedral shape, and the selected bands are nonuniformly spaced. The estimated reduction in the computational complexity is 90% with a 68% reduction in the number of frequency bands. Separation performance achieved during our experimental evaluation was 7.45 (dB) (signal-to-noise ratio) and 2.30 (dB) (cepstral distortion). These results show improvement in performance compared to the use of uniformly spaced frequency band. |
| format | Article |
| id | doaj-art-3094be8187b3475a8fb07c80100c319f |
| institution | Kabale University |
| issn | 2090-0147 2090-0155 |
| language | English |
| publishDate | 2012-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Journal of Electrical and Computer Engineering |
| spelling | doaj-art-3094be8187b3475a8fb07c80100c319f2025-02-03T05:51:14ZengWileyJournal of Electrical and Computer Engineering2090-01472090-01552012-01-01201210.1155/2012/324398324398Practically Efficient Blind Speech Separation Using Frequency Band Selection Based on Magnitude Squared Coherence and a Small Dodecahedral Microphone ArrayKazunobu Kondo0Yusuke Mizuno1Takanori Nishino2Kazuya Takeda3Corporate Research & Development Center, Yamaha Corporation, 203 Matsunokijima, Iwata 438-0192, JapanGraduate School of Engineering, Mie University, 1515 Kurimamachiya-cho, Tsu 514-0102, JapanGraduate School of Information Science, Nagoya University, Chikusa-ku Furou-cho, Nagoya 464-8603, JapanGraduate School of Information Science, Nagoya University, Chikusa-ku Furou-cho, Nagoya 464-8603, JapanSmall agglomerative microphone array systems have been proposed for use with speech communication and recognition systems. Blind source separation methods based on frequency domain independent component analysis have shown significant separation performance, and the microphone arrays are small enough to make them portable. However, the level of computational complexity involved is very high because the conventional signal collection and processing method uses 60 microphones. In this paper, we propose a band selection method based on magnitude squared coherence. Frequency bands are selected based on the spatial and geometric characteristics of the microphone array device which is strongly related to the dodecahedral shape, and the selected bands are nonuniformly spaced. The estimated reduction in the computational complexity is 90% with a 68% reduction in the number of frequency bands. Separation performance achieved during our experimental evaluation was 7.45 (dB) (signal-to-noise ratio) and 2.30 (dB) (cepstral distortion). These results show improvement in performance compared to the use of uniformly spaced frequency band.http://dx.doi.org/10.1155/2012/324398 |
| spellingShingle | Kazunobu Kondo Yusuke Mizuno Takanori Nishino Kazuya Takeda Practically Efficient Blind Speech Separation Using Frequency Band Selection Based on Magnitude Squared Coherence and a Small Dodecahedral Microphone Array Journal of Electrical and Computer Engineering |
| title | Practically Efficient Blind Speech Separation Using Frequency Band Selection Based on Magnitude Squared Coherence and a Small Dodecahedral Microphone Array |
| title_full | Practically Efficient Blind Speech Separation Using Frequency Band Selection Based on Magnitude Squared Coherence and a Small Dodecahedral Microphone Array |
| title_fullStr | Practically Efficient Blind Speech Separation Using Frequency Band Selection Based on Magnitude Squared Coherence and a Small Dodecahedral Microphone Array |
| title_full_unstemmed | Practically Efficient Blind Speech Separation Using Frequency Band Selection Based on Magnitude Squared Coherence and a Small Dodecahedral Microphone Array |
| title_short | Practically Efficient Blind Speech Separation Using Frequency Band Selection Based on Magnitude Squared Coherence and a Small Dodecahedral Microphone Array |
| title_sort | practically efficient blind speech separation using frequency band selection based on magnitude squared coherence and a small dodecahedral microphone array |
| url | http://dx.doi.org/10.1155/2012/324398 |
| work_keys_str_mv | AT kazunobukondo practicallyefficientblindspeechseparationusingfrequencybandselectionbasedonmagnitudesquaredcoherenceandasmalldodecahedralmicrophonearray AT yusukemizuno practicallyefficientblindspeechseparationusingfrequencybandselectionbasedonmagnitudesquaredcoherenceandasmalldodecahedralmicrophonearray AT takanorinishino practicallyefficientblindspeechseparationusingfrequencybandselectionbasedonmagnitudesquaredcoherenceandasmalldodecahedralmicrophonearray AT kazuyatakeda practicallyefficientblindspeechseparationusingfrequencybandselectionbasedonmagnitudesquaredcoherenceandasmalldodecahedralmicrophonearray |