Audio Deepfake Detection
32 papers with code • 1 benchmarks • 3 datasets
Nowadays, deepfake is now generically used by the media or people to refer to any audio or video in which important attributes have been either digitally altered or swapped, with the help of artificial intelligence (AI). Audio deepfake detection is a task that aims to distinguish genuine utterances from fake ones via machine learning techniques.
Benchmarks
These leaderboards are used to track progress in Audio Deepfake Detection
| Trend | Dataset | Best Model | Paper | Code | Compare |
|---|---|---|---|---|---|
|
ASVspoof 2021
SONICS
Most implemented papers
Automatic speaker verification spoofing and deepfake detection using wav2vec 2.0 and data augmentation
takhemlata/ssl_anti-spoofing
•
•
The performance of spoofing countermeasure systems depends fundamentally upon the use of sufficiently representative training data.
AASIST: Audio Anti-Spoofing using Integrated Spectro-Temporal Graph Attention Networks
clovaai/aasist
•
•
Artefacts that differentiate spoofed from bona-fide utterances can reside in spectral or temporal domains.
WaveFake: A Data Set to Facilitate Audio Deepfake Detection
Deep generative modeling has the potential to cause significant harm to society.
WavLM model ensemble for audio deepfake detection
pwc-1/Paper-9
•
•
Audio deepfake detection has become a pivotal task over the last couple of years, as many recent speech synthesis and voice cloning systems generate highly realistic speech samples, thus enabling their use in malicious activities.
Audio Deepfake Detection with Self-Supervised XLS-R and SLS Classifier
qishanzhang/slsforadd
•
•
ACM MM 2024
To enhance the sensitivity of deepfake audio features, we propose a deepfake audio detection model that incorporates an SLS (Sensitive Layer Selection) module.
End-to-end anti-spoofing with RawNet2
eurecom-asp/rawnet2-antispoofing
•
•
Spoofing countermeasures aim to protect automatic speaker verification systems from attempts to manipulate their reliability with the use of spoofed speech signals.
End-to-End Spectro-Temporal Graph Attention Networks for Speaker Verification Anti-Spoofing and Speech Deepfake Detection
eurecom-asp/rawgat-st-antispoofing
•
•
Artefacts that serve to distinguish bona fide speech from spoofed or deepfake speech are known to reside in specific subbands and temporal segments.
Attack Agnostic Dataset: Towards Generalization and Stabilization of Audio DeepFake Detection
piotrkawa/attack-agnostic-dataset
•
•
Audio DeepFakes allow the creation of high-quality, convincing utterances and therefore pose a threat due to its potential applications such as impersonation or fake news.
SpecRNet: Towards Faster and More Accessible Audio DeepFake Detection
In this work, we focus on increasing accessibility to the audio DeepFake detection methods by providing SpecRNet, a neural network architecture characterized by a quick inference time and low computational requirements.
Defense Against Adversarial Attacks on Audio DeepFake Detection
piotrkawa/audio-deepfake-adversarial-attacks
•
•
Audio DeepFakes (DF) are artificially generated utterances created using deep learning, with the primary aim of fooling the listeners in a highly convincing manner.
