rafaelvalle/asrgen
Attacking Speaker Recognition with Deep Generative Models
This project helps security researchers and voice biometric developers explore vulnerabilities in speaker recognition systems. It takes existing audio data and uses deep generative models to create synthetic audio samples that can deceive these systems. The output is 'fake' audio that sounds like a target speaker, useful for evaluating the robustness of voice authentication.
No commits in the last 6 months.
Use this if you need to generate adversarial audio samples to test the security and reliability of speaker recognition or voice biometric systems.
Not ideal if you are looking to build a new speaker recognition system or for general audio synthesis unrelated to security testing.
Stars
34
Forks
6
Language
Jupyter Notebook
License
—
Category
Last pushed
Mar 24, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/rafaelvalle/asrgen"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
kan-bayashi/ParallelWaveGAN
Unofficial Parallel WaveGAN (+ MelGAN & Multi-band MelGAN & HiFi-GAN & StyleMelGAN) with Pytorch
fatchord/WaveRNN
WaveRNN Vocoder + TTS
shangeth/wavencoder
WavEncoder is a Python library for encoding audio signals, transforms for audio augmentation,...
rishikksh20/iSTFTNet-pytorch
iSTFTNet : Fast and Lightweight Mel-spectrogram Vocoder Incorporating Inverse Short-time Fourier...
seungwonpark/melgan
MelGAN vocoder (compatible with NVIDIA/tacotron2)