A Two-Stage Deep Representation Learning-Based Speech Enhancement Method Using Variational Autoencoder and Adversarial Training

TL;DR

This paper introduces a novel two-stage deep learning approach for speech enhancement that combines variational autoencoders and adversarial training to improve speech quality and robustness.

Contribution

It proposes a two-stage DRL-based speech enhancement method using β-VAE and adversarial training to disentangle representations and improve signal estimation accuracy.

Findings

Enhanced speech quality with higher STOI, PESQ, and SI-SDR scores.

Outperforms recent state-of-the-art speech enhancement algorithms.

Demonstrates robustness against inaccurate posterior estimations.

Abstract

This paper focuses on leveraging deep representation learning (DRL) for speech enhancement (SE). In general, the performance of the deep neural network (DNN) is heavily dependent on the learning of data representation. However, the DRL's importance is often ignored in many DNN-based SE algorithms. To obtain a higher quality enhanced speech, we propose a two-stage DRL-based SE method through adversarial training. In the first stage, we disentangle different latent variables because disentangled representations can help DNN generate a better enhanced speech. Specifically, we use the $\beta$-variational autoencoder (VAE) algorithm to obtain the speech and noise posterior estimations and related representations from the observed signal. However, since the posteriors and representations are intractable and we can only apply a conditional assumption to estimate them, it is difficult to ensure that these estimations are always pretty accurate, which may potentially degrade the final accuracy of the signal estimation. To further improve the quality of enhanced speech, in the second stage, we introduce adversarial training to reduce the effect of the inaccurate posterior towards signal reconstruction and improve the signal estimation accuracy, making our algorithm more robust for the potentially inaccurate posterior estimations. As a result, better SE performance can be achieved. The experimental results indicate that the proposed strategy can help similar DNN-based SE algorithms achieve higher short-time objective intelligibility (STOI), perceptual evaluation of speech quality (PESQ), and scale-invariant signal-to-distortion ratio (SI-SDR) scores. Moreover, the proposed algorithm can also outperform recent competitive SE algorithms.