search

sidharthanup / Audio-Analyzer-and-Transcription-Tool

End to end information extraction from audio input.
MIT License
0 stars 2 forks source link

Explore using Whisper with direct files/folders, opposed to string pathnames #26

Closed rphill299 closed 4 months ago

rphill299 commented 5 months ago

In the React UI, we have a file selector. The file selector can be used to upload a file, or a folder (depending on provided settings).

However, it's unclear how to utilize this file (or folder) in the backend for Whisper. We either need to get the full path from the file, or pass Whisper the file instead of the filename.

Apparently this can be done using the librosa python package.