Uncertainty Estimation in Deep Speech Enhancement Using Complex Gaussian Mixture Models

TL;DR

This paper introduces a method to quantify predictive uncertainty in deep speech enhancement by integrating complex Gaussian mixture models, leading to more robust and accurate clean speech estimation.

Contribution

It proposes a novel framework combining statistical CGMMs with neural networks to model the full posterior distribution of clean speech, capturing both aleatoric and epistemic uncertainties.

Findings

Effectively captures predictive uncertainty in speech enhancement.

Achieves superior performance compared to traditional methods.

Demonstrates robustness across multiple datasets.

Abstract

Single-channel deep speech enhancement approaches often estimate a single multiplicative mask to extract clean speech without a measure of its accuracy. Instead, in this work, we propose to quantify the uncertainty associated with clean speech estimates in neural network-based speech enhancement. Predictive uncertainty is typically categorized into aleatoric uncertainty and epistemic uncertainty. The former accounts for the inherent uncertainty in data and the latter corresponds to the model uncertainty. Aiming for robust clean speech estimation and efficient predictive uncertainty quantification, we propose to integrate statistical complex Gaussian mixture models (CGMMs) into a deep speech enhancement framework. More specifically, we model the dependency between input and output stochastically by means of a conditional probability density and train a neural network to map the noisy input to the full posterior distribution of clean speech, modeled as a mixture of multiple complex Gaussian components. Experimental results on different datasets show that the proposed algorithm effectively captures predictive uncertainty and that combining powerful statistical models and deep learning also delivers a superior speech enhancement performance.