dessa-oss/fake-voice-detection
Using temporal convolution to detect Audio Deepfakes
# Technical Summary Builds on ASVSpoof 2019 logical-access dataset with spectrogram preprocessing and includes a pre-trained baseline model (85% accuracy) for immediate inference on custom audio files. Leverages Foundations Atlas for experiment tracking, hyperparameter search automation, and TensorBoard integration to streamline model training and evaluation workflows. Supports distributed job submission and artifact management across GPU-accelerated systems (32GB+ RAM, ≥1 GPU).
383 stars. No commits in the last 6 months.
Stars
383
Forks
86
Language
Python
License
Apache-2.0
Category
Last pushed
Nov 21, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/dessa-oss/fake-voice-detection"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.