-
Notifications
You must be signed in to change notification settings - Fork 11
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
Error message in audio transcription with large files. #89
Comments
Hello, do you have server logs for this error? |
Yes! sorry i forgot to include the log. {"reqId":"t3osM9mo6Tet48DheHS2","level":2,"time":"2024-04-19T13:20:04+00:00","remoteAddr":"","user":"--","app":"integration_openai","method":"","url":"--","message":"API request error : Client error: |
Can you verify your php.ini settings has good values for |
Yes both post_max_size and upload_max_filesize are set to 4G, max_execution_time and max_input_time was 30, i change it to 3600 acording to documentation, restart and still same error. |
The 413 status is coming from localAI I think, see mudler/LocalAI#1733 |
Tnks kyteinsky and marcelklehr for your quick answers. In debug mode localAI does not show me any message about this error but yes it seems to come from it. As exposed in mudler/LocalAI#1733 even whisper-v3 accept larger files but it also seems to have limitations, that lead me to the idea i said before to split the audio files in "chunks", process one by one and concatenate the result, something like: sox orig.wav p.wav silence -l 0 1 0.5 0.1% : newfile : restart #split on silence detection... just an idea! Meanwhile, it will be very usefull a message to the end user in Nextcloud Assistant explaining that it is a limitation. |
Which version of integration_openai are you using?
2.0.0
Which version of Nextcloud are you using?
28.0.2
Which browser are you using? In case you are using the phone App, specify the Android or iOS version and device please.
firefox, chrome
Describe the Bug
Transcribe audio to text produce error if audio file is longer more than 6 min (11 mb aprox). Checking the LocalAI debug there is no prompt request with larger files (11mb). It is seem's an Assistant issue.
Expected Behavior
The expected behavior idealy is to return the complete text. if there is a Assitant limitation show a message saying so. The workaround is to split the original file in "workable" chunks... process it and return the concatenated text of every result.
To Reproduce
Try to trancribe a large audio file more than 10 min.
The text was updated successfully, but these errors were encountered: