Abstract
A generalised Wiener post-filter is proposed to enhance the beamformer output in a microphone array for automatic speech recognition in mobile robot application. Conventional Wiener post-filters are known to perform poorly on suppressing low frequency noise. Proposed is a new post-filter fitting into a generalised Wiener model to reduce low frequency noise more effectively for the purpose of realising robust speech recognition. Experimental results confirm that the proposed post-filter shows significant improvement in terms of speech quality and recognition performance.
| Original language | English |
|---|---|
| Pages (from-to) | 1003-1005 |
| Number of pages | 3 |
| Journal | Electronics Letters |
| Volume | 43 |
| Issue number | 18 |
| DOIs | |
| Publication status | Published - 2007 Sep 6 |
Fingerprint
ASJC Scopus subject areas
- Electrical and Electronic Engineering
Cite this
Microphone array post-filter using input-output ratio of beamformer noise power spectrum. / Yoon, H.; Ko, Hanseok.
In: Electronics Letters, Vol. 43, No. 18, 06.09.2007, p. 1003-1005.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Microphone array post-filter using input-output ratio of beamformer noise power spectrum
AU - Yoon, H.
AU - Ko, Hanseok
PY - 2007/9/6
Y1 - 2007/9/6
N2 - A generalised Wiener post-filter is proposed to enhance the beamformer output in a microphone array for automatic speech recognition in mobile robot application. Conventional Wiener post-filters are known to perform poorly on suppressing low frequency noise. Proposed is a new post-filter fitting into a generalised Wiener model to reduce low frequency noise more effectively for the purpose of realising robust speech recognition. Experimental results confirm that the proposed post-filter shows significant improvement in terms of speech quality and recognition performance.
AB - A generalised Wiener post-filter is proposed to enhance the beamformer output in a microphone array for automatic speech recognition in mobile robot application. Conventional Wiener post-filters are known to perform poorly on suppressing low frequency noise. Proposed is a new post-filter fitting into a generalised Wiener model to reduce low frequency noise more effectively for the purpose of realising robust speech recognition. Experimental results confirm that the proposed post-filter shows significant improvement in terms of speech quality and recognition performance.
UR - http://www.scopus.com/inward/record.url?scp=34548321548&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=34548321548&partnerID=8YFLogxK
U2 - 10.1049/el:20071534
DO - 10.1049/el:20071534
M3 - Article
VL - 43
SP - 1003
EP - 1005
JO - Electronics Letters
JF - Electronics Letters
SN - 0013-5194
IS - 18
ER -