Research

Paper

TESTING March 23, 2026

Disentangling Speaker Traits for Deepfake Source Verification via Chebyshev Polynomial and Riemannian Metric Learning

Authors

Xi Xuan, Wenxin Zhang, Zhiyu Li, Jennifer Williams, Ville Hautamäki, Tomi H. Kinnunen

Abstract

Speech deepfake source verification systems aims to determine whether two synthetic speech utterances originate from the same source generator, often assuming that the resulting source embeddings are independent of speaker traits. However, this assumption remains unverified. In this paper, we first investigate the impact of speaker factors on source verification. We propose a speaker-disentangled metric learning (SDML) framework incorporating two novel loss functions. The first leverages Chebyshev polynomial to mitigate gradient instability during disentanglement optimization. The second projects source and speaker embeddings into hyperbolic space, leveraging Riemannian metric distances to reduce speaker information and learn more discriminative source features. Experimental results on MLAAD benchmark, evaluated under four newly proposed protocols designed for source-speaker disentanglement scenarios, demonstrate the effectiveness of SDML framework. The code, evaluation protocols and demo website are available at https://github.com/xxuan-acoustics/RiemannSD-Net.

Metadata

arXiv ID: 2603.21875
Provider: ARXIV
Primary Category: eess.AS
Published: 2026-03-23
Fetched: 2026-03-24 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.21875v1</id>\n    <title>Disentangling Speaker Traits for Deepfake Source Verification via Chebyshev Polynomial and Riemannian Metric Learning</title>\n    <updated>2026-03-23T12:05:57Z</updated>\n    <link href='https://arxiv.org/abs/2603.21875v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.21875v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Speech deepfake source verification systems aims to determine whether two synthetic speech utterances originate from the same source generator, often assuming that the resulting source embeddings are independent of speaker traits. However, this assumption remains unverified. In this paper, we first investigate the impact of speaker factors on source verification. We propose a speaker-disentangled metric learning (SDML) framework incorporating two novel loss functions. The first leverages Chebyshev polynomial to mitigate gradient instability during disentanglement optimization. The second projects source and speaker embeddings into hyperbolic space, leveraging Riemannian metric distances to reduce speaker information and learn more discriminative source features. Experimental results on MLAAD benchmark, evaluated under four newly proposed protocols designed for source-speaker disentanglement scenarios, demonstrate the effectiveness of SDML framework. The code, evaluation protocols and demo website are available at https://github.com/xxuan-acoustics/RiemannSD-Net.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='eess.AS'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CL'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.SD'/>\n    <published>2026-03-23T12:05:57Z</published>\n    <arxiv:comment>Submitted to Interspeech 2026; The code, evaluation protocols and demo website are available at https://github.com/xxuan-acoustics/RiemannSD-Net</arxiv:comment>\n    <arxiv:primary_category term='eess.AS'/>\n    <author>\n      <name>Xi Xuan</name>\n    </author>\n    <author>\n      <name>Wenxin Zhang</name>\n    </author>\n    <author>\n      <name>Zhiyu Li</name>\n    </author>\n    <author>\n      <name>Jennifer Williams</name>\n    </author>\n    <author>\n      <name>Ville Hautamäki</name>\n    </author>\n    <author>\n      <name>Tomi H. Kinnunen</name>\n    </author>\n  </entry>"
}