Explicit Speech Modeling for Microphone Array Applications

Abstract

In this chapter we address the limitations of current approaches to using microphone arrays for speech acquisition and advocate the development of multi-channel techniques which employ an explicit model of the speech signal. The goal is to combine the advantages of spatial filtering achieved through beamforming with knowledge of the desired time-series attributes. We offer two examples of algorithms which incorporate this principle. The first is a frequency domain approach illustrating the utility of model-based microphone array processing for improved speech spectral analysis. The second is a time domain procedure which highlights the benefits of speech modeling/spatial filtering fusion for the purpose of improving speech quality.