search

SystemSculpt / obsidian-systemsculpt-ai

Enhance your Obsidian App experience with AI-powered tools for note-taking, task management, and much, MUCH more.
MIT License
100 stars 13 forks source link

Split Long Audio To Fit Groq Limits #35

Open alex-ozol opened 3 months ago

alex-ozol commented 3 months ago

Describe the enhancement

Groq Whisper doesn't accept files longer than 25MB for transcription, so it's currently impossible to transcribe such files directly. They can be split into smaller chunks, transcribed one by one, and then all the parts of the transcription can be glued together.

SystemSculpt commented 3 months ago

Ah, thank you for this, the OpenAI version of this already does this, I must have missed the logic apply for Groq's version. Thanks and will have it fixed in next update!