Whisper Configuration #2320
Replies: 2 comments 3 replies
-
Will you use the GPU? Install the IPED python modules (also described in the WIKI): whisperx Another measure would be to download ffmpeg and place it in the path (in the case of WhisperX) |
Beta Was this translation helpful? Give feedback.
-
Download the last 4.2 snapshot: The specific wiki section is this: After installing ONE of the recommended libraries (I recommend faster-whisper if you will run on CPU), go to conf/AudioTranscriptConfig.txt and uncomment the line below: On
The original library doesn't compute confidence scores, we recommend our fork pointed out in the wiki (just if whisperx is used). |
Beta Was this translation helpful? Give feedback.
-
Dear all,
Good afternoon.
I have searched the available topics but could not find a detailed guide for configuring Whisper, specifically for audio transcription. Is there any tutorial or support material on how to set up this solution for transcribing audio files?
Although I am aware that this functionality will be included in the next version, version 4.2 does not yet have a defined release date, and I have an urgent need that needs to be addressed. Therefore, I would like to configure Whisper as soon as possible to meet this requirement.
Thank you for your assistance.
Beta Was this translation helpful? Give feedback.
All reactions