Run Emoti-OpenVoice on your Airbox

This repo is a BM1684X-adapted demo of EmotiVoice(@NetEase Youdao) and OpenVoice(@MyShell.ai).

1. Download bmodels.

• Download the bmodels for EmotiVoice.

python3 -m pip install dfn
python3 -m dfn --url https://disk.sophgo.vip/sharing/KymDuWLGw

• Download bmodels of OpenVoice tone color converter here and run tar zxfv checkpoints.tar.gz.

• The `model_file`` directory should be organized as:

model_file
├── converter
│   ├── checkpoint.pth
│   ├── config.json
│   ├── decoder_1024_f16.bmodel
│   └── decoder_2048_f16.bmodel
├── simbert-base-chinese
│   ├── config.json
│   └── vocab.txt
└── tts
    ├── am_durationpred-1_512_384-1_512_1.bmodel
    ├── am_energypred-1_512_384-1_512_1.bmodel
    ├── am_pitchpred-1_512_384-1_512_1.bmodel
    ├── am_rest_weight.pth
    ├── bert_poolout1-768_1-512_1-512_1-512.bmodel
    ├── hifigan_1-80-1024_F16.bmodel
    ├── jit_am_encoder_1-512-384_1-1-512.bmodel
    └── onnx_am_decoder_1-2048-384.bmodel

2. Run web demo:

sudo apt-get install libsndfile1 -y
pip3 install torch torchaudio
pip3 install numpy numba scipy transformers==4.26.1 librosa soundfile yacs g2p_en jieba pypinyin whisper_timestamped onnxruntime gradio
mv master.zip ~/.cache/torch/hub/
python3 demo_page.py

For more detailed usage, please refer to the README from the original repo of EmotiVoice and OpenVoice.