Retry dictations from history with different models.

[avinashkanaujiya]

I would like to have a feature that allows you to retry already transcribed output with a different model. This would let you test various models with the same input and see how they perform.

[Beingpax]

You can already do this, there is a retranscribe button in the transcript history itself. Just change the selected model from the menu bar or from the UI itself and hit the retranscribe button.

[lenart]

What I would love is to be able to test different models and how they handle transcription without having to generate new power modes for each one of them. The current solution is already useful, but users would definitely benefit from having the transcription models and enhancement models as two separate drop-downs that we could then send my audio file to and see the output.

This is especially useful when testing models against non-English audio files.