Improving Deep Attractor Network by BGRU and GMM for Speech Separation

TL;DR

This paper introduces a simplified and more efficient Deep Attractor Network for speech separation by replacing BLSTM with BGRU and using GMM for clustering, achieving better accuracy and reduced complexity.

Contribution

It proposes a novel DANet model using BGRU and GMM, improving efficiency and accuracy over traditional BLSTM-based DANet.

Findings

Achieved higher SDR and PESQ scores compared to original DANet.

Reduced model complexity by 20.7% and training time by 17.9%.

Performed well on Arabic speech signals, outperforming English results.

Abstract

Deep Attractor Network (DANet) is the state-of-the-art technique in speech separation field, which uses Bidirectional Long Short-Term Memory (BLSTM), but the complexity of the DANet model is very high. In this paper, a simplified and powerful DANet model is proposed using Bidirectional Gated neural network (BGRU) instead of BLSTM. The Gaussian Mixture Model (GMM) other than the k-means was applied in DANet as a clustering algorithm to reduce the complexity and increase the learning speed and accuracy. The metrics used in this paper are Signal to Distortion Ratio (SDR), Signal to Interference Ratio (SIR), Signal to Artifact Ratio (SAR), and Perceptual Evaluation Speech Quality (PESQ) score. Two speaker mixture datasets from TIMIT corpus were prepared to evaluate the proposed model, and the system achieved 12.3 dB and 2.94 for SDR and PESQ scores respectively, which were better than the original DANet model. Other improvements were 20.7% and 17.9% in the number of parameters and time training, respectively. The model was applied on mixed Arabic speech signals and the results were better than that in English.