This repository is forked from Real-Time-Voice-Cloning which only support English.
? Chinese supported mandarin and tested with dataset: aidatatang_200zh
? PyTorch worked for pytorch, tested in version of 1.9.0(latest in August 2021), with GPU Tesla T4 and GTX 2060
? Windows + Linux tested in both Windows OS and linux OS after fixing nits
? Easy & Awesome effect with only newly-trained synthesizer, by reusing the pretrained encoder/vocoder
1. Install Requirements
Follow the original repo to test if you got all environment ready.
**Python 3.7 or higher ** is needed to run the toolbox.
- Install PyTorch.
- Install ffmpeg.
pip install -r requirements.txtto install the remaining necessary packages.
2. Train synthesizer with aidatatang_200zh
Download aidatatang_200zh dataset and unzip: make sure you can access all .wav in train folder
Preprocess with the audios and the mel spectrograms:
python synthesizer_preprocess_audio.py <datasets_root>
Preprocess the embeddings:
python synthesizer_preprocess_embeds.py <datasets_root>/SV2TTS/synthesizer
Train the synthesizer:
python synthesizer_train.py mandarin <datasets_root>/SV2TTS/synthesizer
Go to next step when you see attention line show and loss meet your need in training folder synthesizer/saved_models/.
FYI, my attention came after 18k steps and loss became lower than 0.4 after 50k steps.
3. Launch the Toolbox
You can then try the toolbox:
python demo_toolbox.py -d <datasets_root>
- [x] Add demo video
- [ ] Add support for more dataset
- [ ] Upload pretrained model
- ? Welcome to add more