Comparison of Speech Activity Detection Techniques for Speaker Recognition

TL;DR

This paper systematically reviews various speech activity detection techniques and evaluates their impact on speaker recognition performance under different noise conditions, highlighting the superiority of Gaussian modeling-based SAD.

Contribution

It provides a comparative analysis of SAD techniques in speaker recognition, emphasizing the effectiveness of Gaussian modeling-based methods in noisy environments.

Findings

Gaussian modeling-based SAD outperforms other techniques in noisy conditions

Performance of speaker verification systems depends heavily on SAD choice

Gaussian mixture model-UBM classifier used for evaluation

Abstract

Speech activity detection (SAD) is an essential component for a variety of speech processing applications. It has been observed that performances of various speech based tasks are very much dependent on the efficiency of the SAD. In this paper, we have systematically reviewed some popular SAD techniques and their applications in speaker recognition. Speaker verification system using different SAD technique are experimentally evaluated on NIST speech corpora using Gaussian mixture model- universal background model (GMM-UBM) based classifier for clean and noisy conditions. It has been found that two Gaussian modeling based SAD is comparatively better than other SAD techniques for different types of noises.