jackaduma/CycleGAN-VC2
Voice Conversion by CycleGAN (语音克隆/语音转换): CycleGAN-VC2
Implements improved CycleGAN architecture with two-step adversarial losses, 2-1-2D CNN generators, and Patch GAN discriminators for non-parallel voice conversion without paired training data. Built in PyTorch with preprocessing scripts for mel-cepstral coefficients and fundamental frequency normalization, includes a pretrained model for Mandarin speaker conversion between AISHELL and celebrity voice datasets.
571 stars. No commits in the last 6 months.
Stars
571
Forks
109
Language
Python
License
MIT
Category
Last pushed
Jun 10, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/jackaduma/CycleGAN-VC2"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
coqui-ai/TTS
🐸💬 - a deep learning toolkit for Text-to-Speech, battle-tested in research and production
yeyupiaoling/MASR
Pytorch实现的流式与非流式的自动语音识别框架，同时兼容在线和离线识别，目前支持Conformer、Squeezeformer、DeepSpeech2模型，支持多种数据增强方法。
netease-youdao/EmotiVoice
EmotiVoice 😊: a Multi-Voice and Prompt-Controlled TTS Engine
shivammehta25/Matcha-TTS
[ICASSP 2024] 🍵 Matcha-TTS: A fast TTS architecture with conditional flow matching
keithito/tacotron
A TensorFlow implementation of Google's Tacotron speech synthesis with pre-trained model (unofficial)