Generate OIIAI Cat remix from any input music containing vocal using AI.
python=3.7
tensorflow=2.9
spleeter=2.4
librosa=0.10
soundfile=0.13
Download the repository locally.
Under project directory, run command in terminal python CatSynthesize.py -i {input_audio_file_path}. If input path empty, it is defaulted to train01.wav.
Depending on the tempo and richness of the input music, 1 minute takes about 1-6 minutes to process.
The output will show up in the same directory with the name synthesized-output.wav.
The automatic OIIAI algorithm utilizes two separate AI models: Spleeter for vocal isolation, Vocal-Melody-Extraction for vocal melody extraction.