TY - GEN
T1 - A convolutive spectral decomposition approach to the separation of feedback from target speech
AU - Mysore, Gautham J.
AU - Smaragdis, Paris
PY - 2011
Y1 - 2011
N2 - Feedback is a common problem in teleconferencing systems. Typical usage of an adaptive filter can be effective for feedback reduction but it relies on the presence of such a filter on the side of the far speaker in order to reduce feedback on the side of the near speaker. In order to avoid this reliance on the far speaker's setup, we can use an adaptive filter on the side of the near speaker. Unfortunately, due to non-linear speech coding typically used during speech transmission, these filters perform poorly in this situation. In this paper, we present a novel probabilistic method, using a non-negative convolutive decomposition of spectrogram data to perform feedback reduction by posing the problem as a source separation problem. Our method is robust to non-linear speech coding as well as continuous double-talk, which often presents a challenge to adaptive filters. We compare our method to the use of an adaptive filter and show superior results with respect to standard source separation metrics.
AB - Feedback is a common problem in teleconferencing systems. Typical usage of an adaptive filter can be effective for feedback reduction but it relies on the presence of such a filter on the side of the far speaker in order to reduce feedback on the side of the near speaker. In order to avoid this reliance on the far speaker's setup, we can use an adaptive filter on the side of the near speaker. Unfortunately, due to non-linear speech coding typically used during speech transmission, these filters perform poorly in this situation. In this paper, we present a novel probabilistic method, using a non-negative convolutive decomposition of spectrogram data to perform feedback reduction by posing the problem as a source separation problem. Our method is robust to non-linear speech coding as well as continuous double-talk, which often presents a challenge to adaptive filters. We compare our method to the use of an adaptive filter and show superior results with respect to standard source separation metrics.
KW - Feedback Reduction
KW - Non-Negative Spectrogram Factorization
KW - Source Separation
UR - http://www.scopus.com/inward/record.url?scp=82455212592&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=82455212592&partnerID=8YFLogxK
U2 - 10.1109/MLSP.2011.6064559
DO - 10.1109/MLSP.2011.6064559
M3 - Conference contribution
AN - SCOPUS:82455212592
SN - 9781457716232
T3 - IEEE International Workshop on Machine Learning for Signal Processing
BT - 2011 IEEE International Workshop on Machine Learning for Signal Processing - Proceedings of MLSP 2011
T2 - 21st IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2011
Y2 - 18 September 2011 through 21 September 2011
ER -