A Variance-Preserving Interpolation Approach for Diffusion Models with Applications to Single Channel Speech Enhancement and Recognition

TL;DR

This paper introduces a variance-preserving interpolation diffusion model (VPIDM) that enhances speech enhancement and recognition by reducing steps, improving robustness, and outperforming traditional methods in noisy conditions.

Contribution

The paper presents a novel variance-preserving interpolation diffusion model (VPIDM) that simplifies diffusion processes and improves performance in speech enhancement and recognition tasks.

Findings

VPIDM requires only 25 iterative steps, unlike previous models.

VPIDM outperforms conventional discriminative speech enhancement algorithms.

VPIDM shows increased robustness across different SNR levels.

Abstract

In this paper, we propose a variance-preserving interpolation framework to improve diffusion models for single-channel speech enhancement (SE) and automatic speech recognition (ASR). This new variance-preserving interpolation diffusion model (VPIDM) approach requires only 25 iterative steps and obviates the need for a corrector, an essential element in the existing variance-exploding interpolation diffusion model (VEIDM). Two notable distinctions between VPIDM and VEIDM are the scaling function of the mean of state variables and the constraint imposed on the variance relative to the mean's scale. We conduct a systematic exploration of the theoretical mechanism underlying VPIDM and develop insights regarding VPIDM's applications in SE and ASR using VPIDM as a frontend. Our proposed approach, evaluated on two distinct data sets, demonstrates VPIDM's superior performances over conventional discriminative SE algorithms. Furthermore, we assess the performance of the proposed model under varying signal-to-noise ratio (SNR) levels. The investigation reveals VPIDM's improved robustness in target noise elimination when compared to VEIDM. Furthermore, utilizing the mid-outputs of both VPIDM and VEIDM results in enhanced ASR accuracies, thereby highlighting the practical efficacy of our proposed approach.