Self-supervised learning with diffusion-based multichannel speech enhancement for speaker verification under noisy conditions
Abstract
The paper introduces Diff-Filter, a multichannel speech enhancement approach based on the diffusion probabilistic model, for improving speaker verification performance under noisy and reverberant conditions. It also presents a new two-step training procedure that takes the benefit of self-supervised learning. In the first stage, the Diff-Filter is trained by conducting timedomain speech filtering using a scoring-based diffusion model. In the second stage, the Diff-Filter is jointly optimized with a pre-trained ECAPA-TDNN speaker verification model under a self-supervised learning framework. We present a novel loss based on equal error rate. This loss is used to conduct selfsupervised learning on a dataset that is not labelled in terms of speakers. The proposed approach is evaluated on MultiSV, a multichannel speaker verification dataset, and shows significant improvements in performance under noisy multichannel conditions.
Cite
@article{arxiv.2307.02244,
title = {Self-supervised learning with diffusion-based multichannel speech enhancement for speaker verification under noisy conditions},
author = {Sandipana Dowerah and Ajinkya Kulkarni and Romain Serizel and Denis Jouvet},
journal= {arXiv preprint arXiv:2307.02244},
year = {2023}
}