Dataset · mueller91
Multilingual audio deepfake detection dataset
mueller91/MLAAD
From 100k to 1M multi-language audio recordings for identifying synthesized speech.
What's inside
Inside are audio recordings in English, German, French, Spanish, Ukrainian, Polish, and other languages for audio classification tasks. The dataset is intended for training and checking spoofing and voice deepfake recognition systems. Distributed under a CC-BY-NC 4.0 license strictly for non-commercial scientific research.
What you can do with it
- Training classifiers to detect synthesized voice and deepfakes
- Testing the robustness of voice biometrics systems against spoofing
- Academic research on multilingual AI audio detection
- Tasks
- Audio classification
- Languages
- English, German, French, Spanish, Ukrainian, Polish, Russian, Italian
- Size
- 100K–1M rows
- Downloads this month
- 32.7K
- Likes
- 57
- Access
- on request on Hugging Face
Models trained on it
How to load it
from datasets import load_dataset
ds = load_dataset("mueller91/MLAAD")