Switching Variational Auto-Encoders for Noise-Agnostic Audio-visual Speech Enhancement

TL;DR

This paper introduces Switching Variational Auto-Encoders (SwVAE), which dynamically select between different audio-visual speech enhancement models over time, improving noise-agnostic performance in challenging visual conditions.

Contribution

The paper proposes a novel unsupervised switching VAE framework with a latent Markovian variable to adaptively combine models for noise-agnostic audio-visual speech enhancement.

Findings

SwVAE outperforms traditional models in noisy and cluttered visual conditions.

The model effectively switches between architectures to optimize speech enhancement.

Experimental results demonstrate promising performance of SwVAE.

Abstract

Recently, audio-visual speech enhancement has been tackled in the unsupervised settings based on variational auto-encoders (VAEs), where during training only clean data is used to train a generative model for speech, which at test time is combined with a noise model, e.g. nonnegative matrix factorization (NMF), whose parameters are learned without supervision. Consequently, the proposed model is agnostic to the noise type. When visual data are clean, audio-visual VAE-based architectures usually outperform the audio-only counterpart. The opposite happens when the visual data are corrupted by clutter, e.g. the speaker not facing the camera. In this paper, we propose to find the optimal combination of these two architectures through time. More precisely, we introduce the use of a latent sequential variable with Markovian dependencies to switch between different VAE architectures through time in an unsupervised manner: leading to switching variational auto-encoder (SwVAE). We propose a variational factorization to approximate the computationally intractable posterior distribution. We also derive the corresponding variational expectation-maximization algorithm to estimate the parameters of the model and enhance the speech signal. Our experiments demonstrate the promising performance of SwVAE.