TY - JOUR
T1 - Maximum Likelihood Noise Cancellation Using The Em Algorithm
AU - Feder, Meir
AU - Oppenheim, Alan V.
AU - Weinstein, Ehud
N1 - Funding Information:
Manuscript received June 9, 1987: revised May IO, 1988. This work was supported in part by the Advanced Research Projects Agency monitored by ONR under Contract N00014-81-K-0742 and the National Science Foundation under Grant ECS-8407285 at M.I.T.. and in part by ONR un- der Contract N00014-85-K-0272 at WHOI. M. Feder acknowledges the support of Woods Hole Oceanographic Institution. M. Feder and A. V. Oppenheim are with the Department of Electrical Engineering and Computer Science. Research Laboratory of Electronics. Masrachucetts Institute of Technology. Cambridge. MA 02 139. E. Weinctein is with the Department of Electronic Systems. Tel-Aviv University. Tel-Aliv. Israel. and with the Department of Ocean Engineering. Woods Hole Oceanographic Institution. Woods Hole. MA 02543. IEEE Log Nuniber 8825 I3 I .
PY - 1989/2
Y1 - 1989/2
N2 - Single-microphone speech enhancement systems have typically shown limited performance. Two-microphone systems based on a least-squares error criterion have shown better results in some contexts; however, sometimes the desired (speech) signal is cancelled together with the noise. In this paper we suggest a new approach to the two-microphone speech enhancement problem. Specifically, we formulate a maximum likelihood (ML) problem for estimating the parameters needed for cancelling the noise, and then, we solve this ML problem via the iterative EM (Estimate-Maximize) technique. The resulting procedure shows encouraging results that improve upon the “classical” least-squares approach.
AB - Single-microphone speech enhancement systems have typically shown limited performance. Two-microphone systems based on a least-squares error criterion have shown better results in some contexts; however, sometimes the desired (speech) signal is cancelled together with the noise. In this paper we suggest a new approach to the two-microphone speech enhancement problem. Specifically, we formulate a maximum likelihood (ML) problem for estimating the parameters needed for cancelling the noise, and then, we solve this ML problem via the iterative EM (Estimate-Maximize) technique. The resulting procedure shows encouraging results that improve upon the “classical” least-squares approach.
UR - http://www.scopus.com/inward/record.url?scp=0024615767&partnerID=8YFLogxK
U2 - 10.1109/29.21683
DO - 10.1109/29.21683
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???
AN - SCOPUS:0024615767
SN - 0096-3518
VL - 37
SP - 204
EP - 216
JO - IEEE Transactions on Acoustics, Speech, and Signal Processing
JF - IEEE Transactions on Acoustics, Speech, and Signal Processing
IS - 2
ER -