Option to reprocess the transcript to a higher quality | Voters

Option to reprocess the transcript to a higher quality

closed

Julian

I understand that in order to transcribe the meeting in real-time the transcript is lower quality than as if it was processed as a whole audio session all at once. Perhaps after the meeting I can have the option to re-process the audio from the meeting and get a higher quality transcript?

October 4, 2024

Julian Pscheid

updated the status to

closed

We've made great progress with speech recognition quality in Hedy. At this point the Whisper Large model (available on desktop platforms) and Nemotron models (available everywhere) provide top tier STT results, so a post-processing step would add little, if any, additional value. Closing this request.

this is a very cool feature! particularly for selectively sending audio to any previously enabled server side transcription services

Sridhar

This is one of the use cases for this feature request.

Sebastian

I would like to be able to choose between different levels of detail. Short protocol, detailed content protocol, resolution protocol,...

Autopilot

Merged in a post:

Full accurate transcript at the end of the session

Sebastien R

It would be nice to have a full accurate transcript provided at the end of the session.

January 9, 2025

Julian Pscheid

Maybe process it remotely to enable diarization. Would probably require a new premium subscription level. Will also require option to save the audio of the session.

Julian

updated the status to

under review