google-research / mint

Multi-modal Content Creation Model Training Infrastructure including the FACT model (AI Choreographer) implementation.
Apache License 2.0
497 stars 85 forks source link

_get_tempo function in preprocessing.py might be wrong with m**0.wav files #44

Open Leon0427 opened 2 years ago

Leon0427 commented 2 years ago

I calculate tempo of mBR0, 105, with librosa, which is different to the “int(audio_name[3]) * 10 + 80” code.