Multichannel Speech Enhancement in the Time Domain
Abstract
In this chapter, we show how to extend the ideas of single-channel noise reduction in the time domain to the multichannel (or multisensors) case. The main advantage here is that we take the spatial information into account. This generalization is not straightforward since several steps are required to get results that look similar in their formulation to the single microphone scenario. We show again the importance of the speech subspace in all derived optimal filtering matrix for multichannel speech enhancement.
Keywords
Time domain; Multichannel; Microphone arrays; Orthogonal decomposition; Performance measures; Optimal filtering matrices
In this chapter, we show how to extend ...
Get Speech Enhancement now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
