Unsupervised Multi-channel Speech Dereverberation via Diffusion

TL;DR

This paper introduces USD-DPS, an unsupervised diffusion-based method for multi-channel speech dereverberation that estimates room impulse responses and enforces mixture consistency, achieving superior results without supervised training.

Contribution

The paper proposes a novel unsupervised diffusion model for multi-channel speech dereverberation that jointly estimates RIRs and enforces mixture consistency, advancing beyond prior supervised methods.

Findings

Outperforms existing unsupervised dereverberation methods

Effectively estimates multi-channel RIRs using a combined approach

Achieves superior dereverberation quality in experiments

Abstract

We consider the problem of multi-channel single-speaker blind dereverberation, where multi-channel mixtures are used to recover the clean anechoic speech. To solve this problem, we propose USD-DPS, {U}nsupervised {S}peech {D}ereverberation via {D}iffusion {P}osterior {S}ampling. USD-DPS uses an unconditional clean speech diffusion model as a strong prior to solve the problem by posterior sampling. At each diffusion sampling step, we estimate all microphone channels' room impulse responses (RIRs), which are further used to enforce a multi-channel mixture consistency constraint for diffusion guidance. For multi-channel RIR estimation, we estimate reference-channel RIR by optimizing RIR parameters of a sub-band RIR signal model, with the Adam optimizer. We estimate non-reference channels' RIRs analytically using forward convolutive prediction (FCP). We found that this combination provides a good balance between sampling efficiency and RIR prior modeling, which shows superior performance among unsupervised dereverberation approaches. An audio demo page is provided in https://usddps.github.io/USDDPS_demo/.