Only detecting one speaker for 99% of my audio file when it used to work properly #1817
Unanswered
ColtonBehannon
asked this question in
Q&A
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
I have been using the speaker diarization pipeline for many months now and it has worked great. My project has recently undergone many changes that were not directly related to the speaker identification portion but somehow something must have got messed up along the way and now I cannot identify what has caused the problem. When I upload my test file now it just says SPEAKER_00 for the entire file until one segment near the very end. Previously, it would properly swap back and forth between SPEAKER_00 and SPEAKER_01 several times throughout the file. Any help would be greatly appreciated.
Details:
I am using the speaker diarization pipeline from pretrained:
with diarization_config_path leading me to the config.yaml file (directory and files in image below)
My pyannote package versions are:
Full Code:
Beta Was this translation helpful? Give feedback.
All reactions