FlowFake: Liquid Networks for Audio Deepfake Detection

• URL: http://arxiv.org/abs/2606.19579v1
• Date: Wed, 17 Jun 2026 20:32:32 GMT
• Title: FlowFake: Liquid Networks for Audio Deepfake Detection
• Abstract summary: FlowFake is a Liquid Time-Constant (LTC) architecture with per-neuron adaptive time constants simultaneously resolving spectral (10ms) and prosodic (2s) cues.<n>At only 34K parameters FlowFake achieves formal BIBO stability and O(dt4) integration error.
• Score: 22.027087796412605
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: 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

Related papers

• Audio Deepfake Detection with Half-Truth Localisation Using Cross-Attentive Feature Fusion [0.0]
  CAFNet performs ternary classification (real, fully-fake, or half-truth) and regresses the temporal boundaries of the synthesised region in a single forward pass.<n>On binary detection, it achieves 96.76% accuracy and 3.20% Equal Error Rate (EER), outperforming fine-tuned XLS-R 300M (78.31%) and AST 87M (93.03%) at over 500 times fewer parameters.
  arXiv  Detail & Related papers  (2026-05-28T07:47:22Z)
• Self-Supervised Learning via Flow-Guided Neural Operator on Time-Series Data [57.85958428020496]
  Flow-Guided Neural Operator (FGNO) is a novel framework combining operator learning with flow matching for SSL training.<n>FGNO learns mappings in functional spaces by using Short-Time Fourier Transform to unify different time resolutions.<n>Unlike prior generative SSL methods that use noisy inputs during inference, we propose using clean inputs for representation extraction while learning representations with noise.
  arXiv  Detail & Related papers  (2026-02-12T18:54:57Z)
• End-to-end Audio Deepfake Detection from RAW Waveforms: a RawNet-Based Approach with Cross-Dataset Evaluation [8.11594945165255]
  We propose an end-to-end deep learning framework for audio deepfake detection that operates directly on raw waveforms.<n>Our model, RawNetLite, is a lightweight convolutional-recurrent architecture designed to capture both spectral and temporal features without handcrafted preprocessing.
  arXiv  Detail & Related papers  (2025-04-29T16:38:23Z)
• V2SFlow: Video-to-Speech Generation with Speech Decomposition and Rectified Flow [57.51550409392103]
  We introduce V2SFlow, a novel Video-to-Speech (V2S) framework designed to generate natural and intelligible speech directly from silent talking face videos.<n>To address these challenges, we decompose the speech signal into manageable subspaces, each representing distinct speech attributes, and predict them directly from the visual input.<n>To generate coherent and realistic speech from these predicted attributes, we employ a rectified flow matching decoder built on a Transformer architecture.
  arXiv  Detail & Related papers  (2024-11-29T05:55:20Z)
• Exploring WavLM Back-ends for Speech Spoofing and Deepfake Detection [0.0]
  ASVspoof 5 Challenge Track 1: Speech Deepfake Detection - Open Condition consists of a stand-alone speech deepfake (bonafide vs spoof) detection task. We leverage a pre-trained WavLM as a front-end model and pool its representations with different back-end techniques. Our fused system achieves 0.0937 minDCF, 3.42% EER, 0.1927 Cllr, and 0.1375 actDCF.
  arXiv  Detail & Related papers  (2024-09-08T08:54:36Z)
• YOLO-Stutter: End-to-end Region-Wise Speech Dysfluency Detection [5.42845980208244]
  YOLO-Stutter is a first end-to-end method that detects dysfluencies in a time-accurate manner. VCTK-Stutter and VCTK-TTS simulate natural spoken dysfluencies including repetition, block, missing, replacement, and prolongation.
  arXiv  Detail & Related papers  (2024-08-27T11:31:12Z)
• SceneFake: An Initial Dataset and Benchmarks for Scene Fake Audio Detection [54.74467470358476]
  This paper proposes a dataset for scene fake audio detection named SceneFake. A manipulated audio is generated by only tampering with the acoustic scene of an original audio. Some scene fake audio detection benchmark results on the SceneFake dataset are reported in this paper.
  arXiv  Detail & Related papers  (2022-11-11T09:05:50Z)
• High Fidelity Neural Audio Compression [92.4812002532009]
  We introduce a state-of-the-art real-time, high-fidelity, audio leveraging neural networks. It consists in a streaming encoder-decoder architecture with quantized latent space trained in an end-to-end fashion. We simplify and speed-up the training by using a single multiscale spectrogram adversary.
  arXiv  Detail & Related papers  (2022-10-24T17:52:02Z)
• Fully Automated End-to-End Fake Audio Detection [57.78459588263812]
  This paper proposes a fully automated end-toend fake audio detection method. We first use wav2vec pre-trained model to obtain a high-level representation of the speech. For the network structure, we use a modified version of the differentiable architecture search (DARTS) named light-DARTS.
  arXiv  Detail & Related papers  (2022-08-20T06:46:55Z)
• Audio Deepfake Detection Based on a Combination of F0 Information and Real Plus Imaginary Spectrogram Features [51.924340387119415]
  Experimental results on the ASVspoof 2019 LA dataset show that our proposed system is very effective for the audio deepfake detection task. Our proposed system is very effective for the audio deepfake detection task, achieving an equivalent error rate (EER) of 0.43%, which surpasses almost all systems.
  arXiv  Detail & Related papers  (2022-08-02T02:46:16Z)
• Does Audio Deepfake Detection Generalize? [6.415366195115544]
  We systematize audio spoofing detection by re-implementing and uniformly evaluating architectures from related work. We publish a new dataset consisting of 37.9 hours of found audio recordings of celebrities and politicians, of which 17.2 hours are deepfakes. This may suggest that the community has tailored its solutions too closely to the prevailing ASVSpoof benchmark and that deepfakes are much harder to detect outside the lab than previously thought.
  arXiv  Detail & Related papers  (2022-03-30T12:48:22Z)
• Spotting adversarial samples for speaker verification by neural vocoders [102.1486475058963]
  We adopt neural vocoders to spot adversarial samples for automatic speaker verification (ASV) We find that the difference between the ASV scores for the original and re-synthesize audio is a good indicator for discrimination between genuine and adversarial samples. Our codes will be made open-source for future works to do comparison.
  arXiv  Detail & Related papers  (2021-07-01T08:58:16Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.