Contrastive Knowledge Distillation for Embedding Refinement in Personalized Speech Enhancement

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Serre, Thomas, Fontaine, Mathieu, Benhaim, Éric, Essid, Slim
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911393276493824
author Serre, Thomas
Fontaine, Mathieu
Benhaim, Éric
Essid, Slim
author_facet Serre, Thomas
Fontaine, Mathieu
Benhaim, Éric
Essid, Slim
contents Personalized speech enhancement (PSE) has shown convincing results when it comes to extracting a known target voice among interfering ones. The corresponding systems usually incorporate a representation of the target voice within the enhancement system, which is extracted from an enrollment clip of the target voice with upstream models. Those models are generally heavy as the speaker embedding's quality directly affects PSE performances. Yet, embeddings generated beforehand cannot account for the variations of the target voice during inference time. In this paper, we propose to perform on-thefly refinement of the speaker embedding using a tiny speaker encoder. We first introduce a novel contrastive knowledge distillation methodology in order to train a 150k-parameter encoder from complex embeddings. We then use this encoder within the enhancement system during inference and show that the proposed method greatly improves PSE performances while maintaining a low computational load.
format Preprint
id arxiv_https___arxiv_org_abs_2601_16235
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Contrastive Knowledge Distillation for Embedding Refinement in Personalized Speech Enhancement
Serre, Thomas
Fontaine, Mathieu
Benhaim, Éric
Essid, Slim
Sound
Audio and Speech Processing
Signal Processing
Personalized speech enhancement (PSE) has shown convincing results when it comes to extracting a known target voice among interfering ones. The corresponding systems usually incorporate a representation of the target voice within the enhancement system, which is extracted from an enrollment clip of the target voice with upstream models. Those models are generally heavy as the speaker embedding's quality directly affects PSE performances. Yet, embeddings generated beforehand cannot account for the variations of the target voice during inference time. In this paper, we propose to perform on-thefly refinement of the speaker embedding using a tiny speaker encoder. We first introduce a novel contrastive knowledge distillation methodology in order to train a 150k-parameter encoder from complex embeddings. We then use this encoder within the enhancement system during inference and show that the proposed method greatly improves PSE performances while maintaining a low computational load.
title Contrastive Knowledge Distillation for Embedding Refinement in Personalized Speech Enhancement
topic Sound
Audio and Speech Processing
Signal Processing
url https://arxiv.org/abs/2601.16235