webrtcvad is also needed at the extract_speaker_embed.sh step.
Traceback (most recent call last):
File "extract_speaker_embed.py", line 1, in <module>
from speaker_encoder.voice_encoder import SpeakerEncoder
File "/DYGANVC/speaker_encoder/voice_encoder.py", line 2, in <module>
from speaker_encoder import audio
File "/DYGANVC/speaker_encoder/audio.py", line 6, in <module>
import webrtcvad
ModuleNotFoundError: No module named 'webrtcvad'
webrtcvad is also needed at the extract_speaker_embed.sh step.