An adaptive filter bank based neural network approach for time delay estimation and speech enhancement

TL;DR

This paper introduces an adaptive filter bank neural network approach for accurate time delay estimation and speech enhancement, improving acoustic echo cancellation by combining adaptive filtering with neural network classification.

Contribution

It proposes a novel adaptive filter bank neural network method for TDE and integrates it into an AEC scheme with robust residual echo suppression and gain control.

Findings

Higher performance in TDE accuracy and speech enhancement.

Effective residual echo and noise suppression.

Robustness against varying acoustic conditions.

Abstract

Time delay estimation (TDE) plays a key role in acoustic echo cancellation (AEC) using adaptive filter method. Considerable residual echo will be left if estimation error arises. Here, in this paper, we proposed an adaptive filter bank based neural network approach where the delay is estimated by a bank of adaptive filters with overlapped time scope, and all the energy of filter weights are concatenated and feed to a classification network. The index with maximal probability is chosen as the estimated delay. Based on this TDE, an AEC scheme is designed using a neural network for residual echo and noise suppression, and the optimally-modified log-spectral amplitude (OMLSA) algorithm is adopted to make it robust. Also, a robust automatic gain control (AGC) scheme with spectrum smoothing method is designed to amplify speech segments. Performance evaluations reveal that higher performance can be achieved for our scheme.