-
Notifications
You must be signed in to change notification settings - Fork 1.1k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Faster Whisper sometimes stops running suddenly #620
Comments
Sometimes it stops without error when it's actually "out of memory". |
I am experiencing the same problem. If you are using Windows, it may be related to this issue. This issue does not provide a solution, but it does provide some workarounds. |
Not really, I use Ubuntu with an A100 and 80GB of RAM |
I experience the same issue with a T4 and 16GB of Memory. Suddenly stops and says start processing but them hangs. It does not matter which files I use (have videos with different lengths) multiple container running faster whisper stop randomly. |
me too facing the same problem |
Has anyone encountered this issue and fixed it? |
Yes I have fixed it by checking every video via ffprobe and check if I get a duration. Basically we ignore all videos which would produce an error in ffmpeg and since then our service is super stable. |
I found the simplest way to fix this issue is to clear the CMOS or reset the BIOS. I'm not entirely sure why this works, but recently I messed with the BIOS settings and my PC couldn't boot. After resetting the CMOS, the faster whisper error got fixed too. |
So I have large quantity of short 10-60 seconds audio files. I try to transcribe them with faster whisper in a loop. But sometimes whisper just stops/freezes all of a sudden. I use debug mode but there is no error message, just the last info message that it started processing the file. If I repeat the run it sometimes fails on the same file, sometimes it goes a bit further and fails at another file. I looked into the audio files and they seem to work just fine in any media player. I call faster whisper (large-v1) like this:
The text was updated successfully, but these errors were encountered: