Publication Details
STOPA: A Database of Systematic VariaTion Of DeePfake Audio for Source Tracing and Attribution
Chibber Manasi (University of Eastern Finland)
Mishra Jagabandhu (University of Eastern Finland)
Singh Vishwanath P. (University of Eastern Finland)
Kinnunen Tomi (University of Eastern Finland)
Malinka Kamil, Mgr., Ph.D. (DITS FIT BUT)
source tracing, dataset, anti-spoofing, synthetic speech, deepfake
A key research area in deepfake speech detection is source tracing - determining the origin of synthesised utterances. The approaches may involve identifying the acoustic model (AM), vocoder model (VM), or other generation-specific parameters. However, progress is limited by the lack of a dedicated, systematically curated dataset. To address this, we introduce STOPA, a systematically varied and metadata-rich dataset for deepfake speech source tracing, covering 8 AMs, 6 VMs, and diverse parameter settings across 700k samples from 13 distinct synthesisers. Unlike existing datasets, which often feature limited variation or sparse metadata, STOPA provides a systematically controlled framework covering a broader range of generative factors, such as the choice of the vocoder model, acoustic model, or pretrained weights, ensuring higher attribution reliability. This control improves attribution accuracy, aiding forensic analysis, deepfake detection, and generative model transparency.
@INPROCEEDINGS{FITPUB13384,
author = "Anton Firc and Manasi Chibber and Jagabandhu Mishra and P. Vishwanath Singh and Tomi Kinnunen and Kamil Malinka",
title = "STOPA: A Database of Systematic VariaTion Of DeePfake Audio for Source Tracing and Attribution",
year = 2025,
language = "english",
url = "https://www.fit.vut.cz/research/publication/13384"
}