voice_encoder.py samples_per_frame

resemble-ai / Resemblyzer

A python package to analyze and compare voices with deep learning

Apache License 2.0

2.66k stars 419 forks source link

voice_encoder.py samples_per_frame #61

Open hbo-lambda opened 2 years ago

hbo-lambda commented 2 years ago

I don't understand why the samples_per_frame is samples_per_frame = int((sampling_rate * mel_window_step / 1000))， why not samples_per_frame = int((sampling_rate * mel_window_length / 1000))