Peterz98/DiffspeechResearch

1# Prepare Dataset2 3## LJSpeech 4 5### Download Dataset6```bash7mkdir -p data/raw/ljspeech8cd data/raw9wget https://data.keithito.com/data/speech/LJSpeech-1.1.tar.bz210bzip2 -d LJSpeech-1.1.tar.bz211tar -xvf LJSpeech-1.1.tar12cd ../../13```14 15### Forced Align and Preprocess Dataset16```bash17# Preprocess step: text and unify the file structure.18python data_gen/tts/runs/preprocess.py --config $CONFIG_NAME19# Align step: MFA alignment.20python data_gen/tts/runs/train_mfa_align.py --config $CONFIG_NAME21# Binarization step: Binarize data for fast IO. You only need to rerun this line when running different task if you have `preprocess`ed and `align`ed the dataset before.22python data_gen/tts/runs/binarize.py --config $CONFIG_NAME23```24 25## More datasets will be supported soon...