Evolutionary Multi-Objective Fusion of Deepfake Speech Detectors
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866911560821112832 |
|---|---|
| author | Staněk, Vojtěch Perešíni, Martin Sekanina, Lukáš Firc, Anton Malinka, Kamil |
| author_facet | Staněk, Vojtěch Perešíni, Martin Sekanina, Lukáš Firc, Anton Malinka, Kamil |
| contents | While deepfake speech detectors built on large self-supervised learning (SSL) models achieve high accuracy, employing standard ensemble fusion to further enhance robustness often results in oversized systems with diminishing returns. To address this, we propose an evolutionary multi-objective score fusion framework that jointly minimizes detection error and system complexity. We explore two encodings optimized by NSGA-II: binary-coded detector selection for score averaging and a real-valued scheme that optimizes detector weights for a weighted sum. Experiments on the ASVspoof 5 dataset with 36 SSL-based detectors show that the obtained Pareto fronts outperform simple averaging and logistic regression baselines. The real-valued variant achieves 2.37% EER (0.0684 minDCF) and identifies configurations that match state-of-the-art performance while significantly reducing system complexity, requiring only half the parameters. Our method also provides a diverse set of trade-off solutions, enabling deployment choices that balance accuracy and computational cost. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_01330 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Evolutionary Multi-Objective Fusion of Deepfake Speech Detectors Staněk, Vojtěch Perešíni, Martin Sekanina, Lukáš Firc, Anton Malinka, Kamil Sound Artificial Intelligence Cryptography and Security Machine Learning Neural and Evolutionary Computing While deepfake speech detectors built on large self-supervised learning (SSL) models achieve high accuracy, employing standard ensemble fusion to further enhance robustness often results in oversized systems with diminishing returns. To address this, we propose an evolutionary multi-objective score fusion framework that jointly minimizes detection error and system complexity. We explore two encodings optimized by NSGA-II: binary-coded detector selection for score averaging and a real-valued scheme that optimizes detector weights for a weighted sum. Experiments on the ASVspoof 5 dataset with 36 SSL-based detectors show that the obtained Pareto fronts outperform simple averaging and logistic regression baselines. The real-valued variant achieves 2.37% EER (0.0684 minDCF) and identifies configurations that match state-of-the-art performance while significantly reducing system complexity, requiring only half the parameters. Our method also provides a diverse set of trade-off solutions, enabling deployment choices that balance accuracy and computational cost. |
| title | Evolutionary Multi-Objective Fusion of Deepfake Speech Detectors |
| topic | Sound Artificial Intelligence Cryptography and Security Machine Learning Neural and Evolutionary Computing |
| url | https://arxiv.org/abs/2604.01330 |