Text-based editing with multiple dialogue tracks doesn't work

[Yoel12345]

This has been an issue for me since text-based editing was introduced. I presumed it would be addressed by now since it's such an obvious and seemingly simple bug but it has not, and I haven't found any other posts about it.

The issue is that I'll have multiple dialogue tracks in once sequence (usually this is 2 audio tracks, one isolate for each speaker in an interview), and Premiere will auto-transcribe each clip just fine. But when they are together in the same sequence, only one of the tracks' transcriptions will display in the Text pane. Both tracks have identical settings regarding targeting, sync lock, etc. The only way I can get the dialogue of both speakers to display in the Text pane is to generate a static transcript, which of course comes with all the limitations of static transcripts. Each track contains only one audio file for the entire sequence, though trimming may subdivide it into a few different "clips". The issue exists prior to any such subdivision, and the single-track transcription that is displayed seems unaffected by any subsequent clip splitting.

There is nesting and multicam being used in these projects, but none of the enabled audio tracks are located inside a nested sequence, so IN THEORY it should not be a nesting-related bug. The audio tracks in use are all located in the top-level sequence.

The text-based editing feature is very cool but has been unusable for most of my work without this issue resolved.

Thanks for looking into it.

[Alexander Riss]

Thank you for your feedback.

The way dynamic transcripts currently work is, that they collect the transcript of each audio track and build a unified view of these for the sequence. This happens top down - so if an audio clip is above the other, its transcript will be used instead of the one below. This also means if you have 2 mono mics with an interview, you will only see the top most in the sequence, unless you trim out the parts where the second track is active.

This can be worked around by nesting these two audio tracks in a sequence, and creating a static mix down transcript of these. Now on your editing sequence there will be only one audio channel with a transcript (of the nest) with no ordering conflicts.

Hope this helps!

Alexander

[Yoel12345]

Okay, thank you very much for the clarification. The problem is that one of my last steps is usually to nest that main sequence into shorter "episode" sequences where an intro and outro can easily be tacked on, and the transcript still only shows the one track. So I'm not so sure that's actually working as designed.

I also don't understand why that approach makes any sense. For example, a big part of the reason I want to keep the dialogue separate is because I often need to mute one of the speakers coughing or clearing their throat as I review the footage. By nesting the two tracks, the workflow for that becomes less efficient. Also, I learned to be careful about needless nesting for timeline performance reasons a couple back, though things have certainly improved since then. Perhaps it would make sense to have an alternate mode that allows the transcripts to combine without nesting?

[Kevin-Monahan]

The status of this bug report has been updated.

[Alexander Riss]

Thank you for explaining your workflow. This sounds like there is some potential for additional settings to allow users to have a "mix down" transcript without nesting. The current view was mainly meant to prevent noise from one audio track to cause spurious words to appear in the transcript for another - which would be hard to track down, as in a mix these would not exist.
I will discuss this with the team and see if we can fit this on the road map.

Best regards,
 Alexander

[Yoel12345]

Awesome, thanks for your consideration!