Abstract
This paper describes an algorithm to suppress non-stationary noise as well as stationary noise in a speech enhancement system that employs a two-channel generalized sidelobe canceller (GSC). Our approach builds on recent advances in GSC design involving a transfer function ratio (TFR). The proposed system has four stages. The first stage estimates a new TFR along the acoustic paths from the non-stationary noise source to the microphones and the power of the stationary noise components. Second, the estimated power of the stationary noise components is used to execute spectral subtraction (SS) with respect to the input signals. Thirdly, the optimal gain is estimated for speech enhancement on the primary channel. In the final stage, an adaptive filter reduces the residual correlated noise components of the signal. These algorithmic improvements consistently give a better performance than a transfer function based GSC (TF-GSC) alone or a GSC with SS post-filtering under various noise conditions while slightly increasing the computational complexity.
| Original language | English |
|---|---|
| Pages (from-to) | 521-533 |
| Number of pages | 13 |
| Journal | Speech Communication |
| Volume | 51 |
| Issue number | 6 |
| DOIs | |
| Publication status | Published - 2009 Jun 1 |
Fingerprint
Keywords
- Adaptive signal processing
- Generalized sidelobe canceller
- Non-stationary noise
- Speech enhancement
- Transfer function ratio
ASJC Scopus subject areas
- Modelling and Simulation
- Computer Science Applications
- Computer Vision and Pattern Recognition
- Software
- Communication
- Linguistics and Language
Cite this
Extension of two-channel transfer function based generalized sidelobe canceller for dealing with both background and point-source noise. / Kim, Kihyeon; Baran, Robert H.; Ko, Hanseok.
In: Speech Communication, Vol. 51, No. 6, 01.06.2009, p. 521-533.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Extension of two-channel transfer function based generalized sidelobe canceller for dealing with both background and point-source noise
AU - Kim, Kihyeon
AU - Baran, Robert H.
AU - Ko, Hanseok
PY - 2009/6/1
Y1 - 2009/6/1
N2 - This paper describes an algorithm to suppress non-stationary noise as well as stationary noise in a speech enhancement system that employs a two-channel generalized sidelobe canceller (GSC). Our approach builds on recent advances in GSC design involving a transfer function ratio (TFR). The proposed system has four stages. The first stage estimates a new TFR along the acoustic paths from the non-stationary noise source to the microphones and the power of the stationary noise components. Second, the estimated power of the stationary noise components is used to execute spectral subtraction (SS) with respect to the input signals. Thirdly, the optimal gain is estimated for speech enhancement on the primary channel. In the final stage, an adaptive filter reduces the residual correlated noise components of the signal. These algorithmic improvements consistently give a better performance than a transfer function based GSC (TF-GSC) alone or a GSC with SS post-filtering under various noise conditions while slightly increasing the computational complexity.
AB - This paper describes an algorithm to suppress non-stationary noise as well as stationary noise in a speech enhancement system that employs a two-channel generalized sidelobe canceller (GSC). Our approach builds on recent advances in GSC design involving a transfer function ratio (TFR). The proposed system has four stages. The first stage estimates a new TFR along the acoustic paths from the non-stationary noise source to the microphones and the power of the stationary noise components. Second, the estimated power of the stationary noise components is used to execute spectral subtraction (SS) with respect to the input signals. Thirdly, the optimal gain is estimated for speech enhancement on the primary channel. In the final stage, an adaptive filter reduces the residual correlated noise components of the signal. These algorithmic improvements consistently give a better performance than a transfer function based GSC (TF-GSC) alone or a GSC with SS post-filtering under various noise conditions while slightly increasing the computational complexity.
KW - Adaptive signal processing
KW - Generalized sidelobe canceller
KW - Non-stationary noise
KW - Speech enhancement
KW - Transfer function ratio
UR - http://www.scopus.com/inward/record.url?scp=63649133274&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=63649133274&partnerID=8YFLogxK
U2 - 10.1016/j.specom.2009.02.001
DO - 10.1016/j.specom.2009.02.001
M3 - Article
VL - 51
SP - 521
EP - 533
JO - Speech Communication
JF - Speech Communication
SN - 0167-6393
IS - 6
ER -