Ability to change output/input note language after recoding
S
Simona Manescu
I work in multiple languages, and sometimes I forget to update the recording’s input language before uploading the audio transcript to generate a note. When I upload audio with the wrong language selected, it understandably results in an error. It would be very helpful if there were an option to change the input language after uploading the audio, so we don't need to delete the session and restart the entire process. Thank you! :)
Photo Viewer
View photos in a modal
Heidi Team
Thanks so much for sharing this — and for the clear example. Totally get how frustrating it is to have to delete a session and start over when you’re switching languages a few times a week.
Being able to change the
input
language after an upload (and then re-run the transcript/note) makes a lot of sense. I’ll share this request with the team.A couple quick questions so we capture it properly:
1) After switching the input language, would you expect Heidi to re-transcribe from the original audio automatically, or just use the existing transcript?
2) Do you usually want the
output
note language to stay the same (e.g., always English), even when the input language changes?3) Does this happen mostly with uploaded audio files, or also with live recordings?
If you can share one of the error messages you see (or a screenshot like the one above), that’d help us understand the exact failure state too.
S
Simona Manescu
Heidi Team
Hi thank you so much for your attention to this request! It is very appreciated!
1) Well in this particular case, I would expect Heidi to automatically re-transcribe from the original audio since no existing transcript is available (the error occurred due to mismatching languages).
2) From my point of view, It would make the most sense for the output note language to match the output language set at upload. That said, it would be great to also be able to change the language at the individual note/document level. For example: if I upload audio in English and set the output to French, all documents and notes are generated in French — which works well in most cases. However, if a client needs a report in French but a letter in English, note-level language control would provide that flexibility. In short, I see language operating at three levels:
1. Audio input language
2. Output language (set at upload)
3. Individual document/note output language
3) I don't use live recordings, as I prefer to keep an audio/transcript backup; so unfortunately I don't know how or if it happens with live recordings.
As for any additional errors, I only encounter "Sorry, we couldn't generate the document". But again, the error does make sense given the language mismatch. So I don't see it as an issue.
Thank you again for looking into it, I truly appreciated.