<img src="img/emo-tts.png" style="float: left; margin-right: 0px;" />
conda create -n emo python=3.8
conda activate emo
pip install -r requirements.txt
python env.py
Download the model by this link, and then put them into /chinese-roberta-wwm-ext
Collet the data by this
Use this code to complete the following preprocessing:
# The audio path and corresponding text and emotion are stored and divided into training set and validation set.
python getdata.py
python split.py
cd monotonic_align
python setup.py build_ext --inplace
cd ..
python train.py -c path/to/json -m model
python infer.py