Multi-Sinkhorn Teacher Knowledge Aggregation Framework for Adaptive Audio Anti-Spoofing

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F26%3A0199981" target="_blank" >RIV/00216305:26230/26:0199981 - isvavai.cz</a>

Výsledek na webu

<a href="https://ieeexplore.ieee.org/abstract/document/11150711" target="_blank" >https://ieeexplore.ieee.org/abstract/document/11150711</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/TASLPRO.2025.3606191" target="_blank" >10.1109/TASLPRO.2025.3606191</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Multi-Sinkhorn Teacher Knowledge Aggregation Framework for Adaptive Audio Anti-Spoofing

Popis výsledku v původním jazyce

Audio anti-spoofing algorithms are widely deployed to defend against spoofing attacks, yet they often fail to detect unseen attacks. Although unsupervised domain adaptation (UDA) offers the potential to address this challenge, existing methods struggle with the large intra-class variability and complex distribution structures in target domains caused by the diversity of speech and attack types. In contrast, optimal transport (OT) leverages the geometric structure of intra-class distributions to measure discrepancies between probability distributions. The effectiveness of OT relies on the discriminability of data within target domains. However, in real-world scenarios involving multiple target domains, these domains often overlap in feature space, leading to the negative transport problem in OT. To overcome these domain mismatches in anti-spoofing, we propose the Multi-Sinkhorn Teacher Knowledge Aggregation (MSTKA) framework. Initially, to avoid interference between target domains during alignment, we use OT to adapt the source model to each target domain independently, thereby reducing negative transport. This adaptation involves constructing an OT cost matrix based on sentence-level representations of cross-domain samples and training an expert model for each target domain. Subsequently, we aggregate the knowledge from these expert models into a unified student model, enabling it to generalize across multiple target domains. Since spoofing cues could be distributed across different temporal scales, we align the student model's representations at multiple time scales with the teacher model's sentence-level representations to enhance the effectiveness of knowledge distillation. Multi-target adaptation experiments on eleven data sets demonstrate that our framework achieves state-of-the-art performance in audio anti-spoofing.

Název v anglickém jazyce

Multi-Sinkhorn Teacher Knowledge Aggregation Framework for Adaptive Audio Anti-Spoofing

Popis výsledku anglicky

Audio anti-spoofing algorithms are widely deployed to defend against spoofing attacks, yet they often fail to detect unseen attacks. Although unsupervised domain adaptation (UDA) offers the potential to address this challenge, existing methods struggle with the large intra-class variability and complex distribution structures in target domains caused by the diversity of speech and attack types. In contrast, optimal transport (OT) leverages the geometric structure of intra-class distributions to measure discrepancies between probability distributions. The effectiveness of OT relies on the discriminability of data within target domains. However, in real-world scenarios involving multiple target domains, these domains often overlap in feature space, leading to the negative transport problem in OT. To overcome these domain mismatches in anti-spoofing, we propose the Multi-Sinkhorn Teacher Knowledge Aggregation (MSTKA) framework. Initially, to avoid interference between target domains during alignment, we use OT to adapt the source model to each target domain independently, thereby reducing negative transport. This adaptation involves constructing an OT cost matrix based on sentence-level representations of cross-domain samples and training an expert model for each target domain. Subsequently, we aggregate the knowledge from these expert models into a unified student model, enabling it to generalize across multiple target domains. Since spoofing cues could be distributed across different temporal scales, we align the student model's representations at multiple time scales with the teacher model's sentence-level representations to enhance the effectiveness of knowledge distillation. Multi-target adaptation experiments on eleven data sets demonstrate that our framework achieves state-of-the-art performance in audio anti-spoofing.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

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

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2025

Kód důvěrnosti údajů

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

Název periodika

IEEE Transactions on Audio, Speech, and Language Processing

ISSN

1558-7916

e-ISSN

1558-7924

Svazek periodika

—

Číslo periodika v rámci svazku

33

Stát vydavatele periodika

US - Spojené státy americké

Počet stran výsledku

16

Strana od-do

3850-3865

Kód UT WoS článku

001579024300004

EID výsledku v databázi Scopus

—