Phase Aware Speech Enhancement using Realisation of Complex-valued LSTM

TL;DR

This paper introduces a complex-valued LSTM network for speech enhancement that effectively models phase information, leading to improved speech quality over traditional magnitude-based methods.

Contribution

It proposes a novel RCLSTM architecture for complex ratio mask estimation, capturing sequential dependencies and phase information in speech enhancement.

Findings

RCLSTM outperforms real-valued masking methods in objective measures.

The method improves PESQ scores by over 4.3%.

Effective preservation of phase information enhances speech quality.

Abstract

Most of the deep learning based speech enhancement (SE) methods rely on estimating the magnitude spectrum of the clean speech signal from the observed noisy speech signal, either by magnitude spectral masking or regression. These methods reuse the noisy phase while synthesizing the time-domain waveform from the estimated magnitude spectrum. However, there have been recent works highlighting the importance of phase in SE. There was an attempt to estimate the complex ratio mask taking phase into account using complex-valued feed-forward neural network (FFNN). But FFNNs cannot capture the sequential information essential for phase estimation. In this work, we propose a realisation of complex-valued long short-term memory (RCLSTM) network to estimate the complex ratio mask (CRM) using sequential information along time. The proposed RCLSTM is designed to process the complex-valued sequences using complex arithmetic, and hence it preserves the dependencies between the real and imaginary parts of CRM and thereby the phase. The proposed method is evaluated on the noisy speech mixtures formed from the Voice-Bank corpus and DEMAND database. When compared to real value based masking methods, the proposed RCLSTM improves over them in several objective measures including perceptual evaluation of speech quality (PESQ), in which it improves by over 4.3%