cs.SD · 2607.04848 Copy arXiv ID · Jul 6, 2026 Save SynSFX: Multi-Model Sound Effects Synthesis Dataset for Deepfake Detection and Evaluation Authors: Linxi Li , Yuncong Yu , Qianwei Guo , Liwei Jin , Yechen Wang , Carsten Maple
Abstract While audio deepfake detection has advanced significantly, representative detectors show limited generalization to synthetic sound effects. Existing environmental audio datasets such as EnvSDD provide important initial resources, but remain limited in scale and generation provenance for studying isolated sound-effect deepfakes. To support this direction, we present SynSFX, a large-scale corpus of 43374 clips (26452 synthetic, 16922 real) spanning 7 popular text-to-audio models.
Explore similar work Jun 17, 2026 · Shivaay Dhondiyal, Divyansh Sharma, Dinesh Kumar Vishwakarma Audio Deepfake Detection Fake News
Apr 21, 2026 · Khoi Vu, Dat Tran, Khanh Do +8 Audio Deepfake Detection Sonification
Mar 24, 2026 · Octavian Pascu, Dan Oneata, Horia Cucu +1 Audio Deepfake Detection Music Generation
Jun 17, 2026 · cs.SD J/K move · Enter open · S save
Shivaay Dhondiyal, Divyansh Sharma, Dinesh Kumar Vishwakarma
Delhi Technological University, New Delhi, India.
Audio deepfakes generated by neural text-to-speech and voice-cloning systems threaten speaker verification and public discourse at scale. The core challenge is cross-dataset generalization: detectors trained on one synthesis pipeline collapse on unseen forgeries. We argue that this failure is primarily because of structural synthetic speech artifacts which are multi-timescale trajectory anomalies. Though every existing detector aggregates a fixed-window frame statistics, this misaligns the architecture with the signal. We propose FlowFake, a Liquid Time-Constant (LTC) architecture whose hidden state evolves via a learned ODE, with per-neuron adaptive time constants simultaneously resolving spectral (10ms) and prosodic (2s) cues. At only 34K parameters FlowFake achieves formal BIBO stability and O(dt^4) integration error. On a four-dataset cross domain benchmark (ASVspoof2019-LA, FakeOrReal, InTheWild, MLAAD), FlowFake reaches 75.29% on ASVspoof2019 trained only on FakeOrReal and 79.97% trained only on MLAAD. It outperforms RawGAT-ST and Whisper-DF on every evaluated pair and matching SSL Wav2vec2 (300x larger) at 0.01% of its parameter count. The source code is available on : https://github.com/GhostRider2023/FlowFake