Detection Of Cross-dataset Fake Audio Based On Prosodic And Pronunciation Features

Abstract

Existing fake audio detection systems perform well in in-domain testing, but still face many challenges in out-of-domain testing. This is due to the mismatch between the training and test data, as well as the poor generalizability of features extracted from limited views. To address this, we propose multi-view features for fake audio detection, which aim to capture more generalized features from prosodic, pronunciation, and wav2vec dimensions. Specifically, the phoneme duration features are extracted from a pre-trained model based on a large amount of speech data. For the pronunciation features, a Conformer-based phoneme recognition model is first trained, keeping the acoustic encoder part as a deeply embedded feature extractor. Furthermore, the prosodic and pronunciation features are fused with wav2vec features based on an attention mechanism to improve the generalization of fake audio detection models. Results show that the proposed approach achieves significant performance gains in

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keywang2023detection

Related papers

• Continual Learning For Fake Audio Detection (2021)11.49
• Pitch Imperfect: Detecting Audio Deepfakes Through Acoustic Prosodic Analysis (2025)0.00
• Mixture Of Experts Fusion For Fake Audio Detection Using Frozen Wav2vec 2.0 (2024)0.00
• Training-free Deepfake Voice Recognition By Leveraging Large-scale Pre-trained Models (2024)9.23
• AUDETER: A Large-scale Dataset For Deepfake Audio Detection In Open Worlds (2025)0.00
• Self-attention And Hybrid Features For Replay And Deep-fake Audio Detection (2024)0.00
• Xmad-bench: Cross-domain Multilingual Audio Deepfake Benchmark (2025)1.69
• Combining Automatic Speaker Verification And Prosody Analysis For Synthetic Speech Detection (2022)10.48