Methods for noise cancellation based on the EM algorithm
Meir Feder, Alan V. Oppenheim, Ehud Weinstein
Abstract
Meir Feder, Alan V. Oppenheim, Ehud Weinstein
Abstract
Single microphone speech enhancement systems have typically shown limited performance, while multiple microphone systems based on a least-squares error criterion have shown encouraging results in some contexts. In this paper we formulate a new approach to multiple microphone speech enhancement. Specifically, we formulate a maximum likelihood (ML) problem for estimating the parameters needed for canceling the noise in a two microphone speech enhancement system. This ML problem is solved via the iterative EM (Estimate-Maximize) technique. The resulting algorithm shows encouraging results when applied to the speech enhancement problem.
OpenAlex reports 15 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Single microphone speech enhancement systems have typically shown limited performance, while multiple microphone systems based on a least-squares error criterion have shown encouraging results in some contexts. In this paper we formulate a new approach to multiple microphone speech enhancement. Specifically, we formulate a maximum likelihood (ML) problem for estimating the parameters needed for canceling the noise in a two microphone speech enhancement system. This ML problem is solved via the iterative EM (Estimate-Maximize) technique. The resulting algorithm shows encouraging results when applied to the speech enhancement problem.
Key concepts: Microphone, Speech enhancement, Computer science, Noise (video), Speech recognition, Active noise control, Algorithm, Iterative method