Saved in:
Bibliographic Details
Main Authors: Xuan, Xi, Zhu, Zimo, Zhang, Wenxin, Lin, Yi-Cheng, Kinnunen, Tomi
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.09294
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913987684204544
author Xuan, Xi
Zhu, Zimo
Zhang, Wenxin
Lin, Yi-Cheng
Kinnunen, Tomi
author_facet Xuan, Xi
Zhu, Zimo
Zhang, Wenxin
Lin, Yi-Cheng
Kinnunen, Tomi
contents Advances in speech synthesis intensify security threats, motivating real-time deepfake detection research. We investigate whether bidirectional Mamba can serve as a competitive alternative to Self-Attention in detecting synthetic speech. Our solution, Fake-Mamba, integrates an XLSR front-end with bidirectional Mamba to capture both local and global artifacts. Our core innovation introduces three efficient encoders: TransBiMamba, ConBiMamba, and PN-BiMamba. Leveraging XLSR's rich linguistic representations, PN-BiMamba can effectively capture the subtle cues of synthetic speech. Evaluated on ASVspoof 21 LA, 21 DF, and In-The-Wild benchmarks, Fake-Mamba achieves 0.97%, 1.74%, and 5.85% EER, respectively, representing substantial relative gains over SOTA models XLSR-Conformer and XLSR-Mamba. The framework maintains real-time inference across utterance lengths, demonstrating strong generalization and practical viability. The code is available at https://github.com/xuanxixi/Fake-Mamba.
format Preprint
id arxiv_https___arxiv_org_abs_2508_09294
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Fake-Mamba: Real-Time Speech Deepfake Detection Using Bidirectional Mamba as Self-Attention's Alternative
Xuan, Xi
Zhu, Zimo
Zhang, Wenxin
Lin, Yi-Cheng
Kinnunen, Tomi
Audio and Speech Processing
Artificial Intelligence
Computation and Language
Machine Learning
Systems and Control
Advances in speech synthesis intensify security threats, motivating real-time deepfake detection research. We investigate whether bidirectional Mamba can serve as a competitive alternative to Self-Attention in detecting synthetic speech. Our solution, Fake-Mamba, integrates an XLSR front-end with bidirectional Mamba to capture both local and global artifacts. Our core innovation introduces three efficient encoders: TransBiMamba, ConBiMamba, and PN-BiMamba. Leveraging XLSR's rich linguistic representations, PN-BiMamba can effectively capture the subtle cues of synthetic speech. Evaluated on ASVspoof 21 LA, 21 DF, and In-The-Wild benchmarks, Fake-Mamba achieves 0.97%, 1.74%, and 5.85% EER, respectively, representing substantial relative gains over SOTA models XLSR-Conformer and XLSR-Mamba. The framework maintains real-time inference across utterance lengths, demonstrating strong generalization and practical viability. The code is available at https://github.com/xuanxixi/Fake-Mamba.
title Fake-Mamba: Real-Time Speech Deepfake Detection Using Bidirectional Mamba as Self-Attention's Alternative
topic Audio and Speech Processing
Artificial Intelligence
Computation and Language
Machine Learning
Systems and Control
url https://arxiv.org/abs/2508.09294