Original Authors : Jesse Engel, Lamtharn (Hanoi) Hantrakul, Chenjie Gu, Adam Roberts (Google)

This Repository is NOT an official implement of authors.

• Link

1. Clone this repository

git clone https://github.com/sweetcocoa/ddsp-pytorch

2. Prepare your own audio data. (wav, mp3, flac.. )
3. Use ffmpeg to convert that audio's sampling rate to 16k

# example
ffmpeg -y -loglevel fatal -i $input_file -ac 1 -ar 16000 $output_file

4. Use CREPE to precalculate the fundamental frequency of the audio.

# example
crepe directory-to-audio/ --output directory-to-audio/f0_0.004/  --viterbi --step-size 4

5. MAKE config file. (See configuration config/violin.yaml to make appropriate config file.) And edit train/train.py

config = setup(default_config="../configs/your_config.yaml")

6. Run train/train.py

cd train
python train.py

cd train
python test.py\ 
--input input.wav\
--output output.wav\
--ckpt trained_weight.pth\
--config config/your-config.yaml\
--wave_length 16000

download

• Jongho Choi ([email protected], BS Student @ Seoul National Univ.)
• Sungho Lee ([email protected], BS Student @ Postech.)

Equally contributed.