Performance evaluation of beamforming techniques for speech enhancement

Abstract

This paper presents speech enhancement using fixed and adaptive beamforming techniques to improve quality of speech signal in noisy environment. Microphone arrays provide a means of enhancing a desired signal in the presence of corrupting noise sources using spatial filtering. In this paper, Delay and Sum Beamformer (DSB) and Generalized Side lobe Canceller (GSC) beamformer are implemented using microphone array and their performance are evaluated by considering various noises under different SNR levels. Enhanced speech characteristics are represented in time domain and frequency domain. The interference is nullified and the beamformer is steered towards desired direction using spatial response. An enhanced speech with improved SNR is achieved by generalized side lobe canceller with Least Mean Square (LMS) algorithm than Delay and Sum beamformer.