Pushing the Performance of Synthetic Speech Detection with Kolmogorov-Arnold Networks and Self-Supervised Learning Models

Tuan Dat Phuong, Long-Vu Hoang, Huy Dat Tran
6/17/2025
cs.SDcs.CLeess.AS

No official code found. Best implementation: HuSTeP-Human-Speech-Text-Processing-Lab/XLSR-GRKAN-Conformer (2 stars, MIT, last commit 2025-08-18).

Abstract

Recent advancements in speech synthesis technologies have led to increasingly advanced spoofing attacks, posing significant challenges for automatic speaker verification systems. While systems based on self-supervised learning (SSL) models, particularly the XLSR-Conformer model, have demonstrated remarkable performance in synthetic speech detection, there remains room for architectural improvements. In this paper, we propose a novel approach that replaces the traditional Multi-Layer Perceptron in the XLSR-Conformer model with a Kolmogorov-Arnold Network (KAN), a novel architecture based on the Kolmogorov-Arnold representation theorem. Our results on ASVspoof2021 demonstrate that integrating KAN into the SSL-based models can improve the performance by 60.55% relatively on LA and DF sets, further achieving 0.70% EER on the 21LA set. These findings suggest that incorporating KAN into SSL-based models is a promising direction for advances in synthetic speech detection.

View on arXivView PDF

Code Implementations(1)

HuSTeP-Human-Speech-Text-Processing-Lab/XLSR-GRKAN-ConformerCommunity100%
MIT

Official implementation of the paper "Pushing the Performance of Synthetic Speech Detection with Kolmogorov-Arnold Networks and Self-Supervised Learning Models", accepted at INTERSPEECH 2025.

20PythonMay 24, 20251 years agoMIT

Similar Papers

Attentive Merging of Hidden Embeddings from Pre-trained Speech Model for Anti-spoofing Detection

82% match1 repo92024

Self-supervised Neural Factor Analysis for Disentangling Utterance-level Speech Representations

82% match2023

Collaborative Watermarking for Adversarial Speech Synthesis

82% match2023

HuLA: Prosody-Aware Anti-Spoofing with Multi-Task Learning for Expressive and Emotional Synthetic Speech

81% match1 repo22025

Efficient infusion of self-supervised representations in Automatic Speech Recognition

81% match2024

"KAN you hear me?" Exploring Kolmogorov-Arnold Networks for Spoken Language Understanding

81% match1 repo12025

Cite this paper

@article{phuong2025pushing,
  title  = {Pushing the Performance of Synthetic Speech Detection with Kolmogorov-Arnold Networks and Self-Supervised Learning Models},
  author = {Tuan Dat Phuong and Long-Vu Hoang and Huy Dat Tran},
  year   = {2025},
  eprint = {2506.14153},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2506.14153v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2506.14153.svg)](https://codeofpaper.com/paper/2506.14153)

Discussion