search

sidharthanup / Audio-Analyzer-and-Transcription-Tool

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

Finish the whisper-transcribe-folder endpoint #27

Closed rphill299 closed 4 months ago

rphill299 commented 5 months ago

I think whether we select a folder or some number of files, the batched endpoint should work, so we should delete the whisper-transcribe-folder endpoint and replace with a call to the batched transcription endpoint

rphill299 commented 4 months ago

After some small updates, the folder endpoint should now work with batched processing of strings. It works on the sequential version, so it should work on the batched GPU version (needs Sid to test)

rphill299 commented 4 months ago

Just note, the "Input folder" string must match the folder you select!! if you select {path}/outputs as your input folder, then your string should be "{path}/outputs"