Continual Unsupervised Domain Adaptation for Audio Deepfake Detection

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F26%3A0199990" target="_blank" >RIV/00216305:26230/26:0199990 - isvavai.cz</a>

Result on the web

<a href="https://ieeexplore.ieee.org/document/10890538" target="_blank" >https://ieeexplore.ieee.org/document/10890538</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/ICASSP49660.2025.10890538" target="_blank" >10.1109/ICASSP49660.2025.10890538</a>

Result language

angličtina

Original language name

Continual Unsupervised Domain Adaptation for Audio Deepfake Detection

Original language description

Audio deepfake detection (ADD) aims to verify the authenticity of audio. However, its performance declines sharply when facing significant domain discrepancies caused by unknown datasets. Unsupervised domain adaptation (UDA) has been applied to mitigate domain mismatch. However, as generative models evolve, existing UDA methods struggle with catastrophic forgetting when facing continuously emerging spoofing methods. To address this challenge, we introduce continual UDA for ADD, which involves sequentially training across multiple target domains with continual learning. We propose a causality-distillation-based continual domain adversarial training framework for continual UDA, called CD-DAT. Specifically, we employ the domain adversarial training (DAT) framework to learn both spoofing-discriminative and domain-invariant deep features. In addition, we design a continual learning algorithm utilizing causality distillation to capture the mapping between utterances and classes, effectively mitigating forgetting and maintaining generalization. Experiments demonstrated that CD-DAT improved detection performance across all domains, confirming its memory stability and learning plasticity.

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

—

Continuities

S - Specificky vyzkum na vysokych skolach

Publication year

2025

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection

Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing

ISBN

979-8-3503-6874-1

ISSN

—

e-ISSN

—

Number of pages

5

Pages from-to

—

Publisher name

Institute of Electrical and Electronics Engineers Inc.

Place of publication

Hyderabad, Indická republika

Event location

Hyderabad, Indická republika

Event date

Apr 6, 2025

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

—