As we drive toward release of PPro’s initial UXP API support, we’re beginning the planning process for API feature requests (beyond “make sure UXP can do what CEP+ExtendScript used to be able to do”).
Toward that end…
What Captions-related APIs would you need, to support workflows desired by your users?
Any CRUD for caption tracks, and items, applying different styles, and moving them around different tracks.
Would be useful to be able to create new styles.
Would be really useful to use data from the transcript, I know a lot of devs have to generate their own transcript for their tools, so it’d be nice to be able to use the existing data.
Would also be nice to be able to convert captions to graphics.
I think I speak for both developers and end users when I say that the overarching desire is for accurate per-word timing in the Caption track, Graphics Text layers, MOGRTs, and independent plugins.
Developers currently have hacky workarounds and make use of external tools. The most common workflow is to use a cloud-based service (Whisper, etc) to generate transcript data and then spit PNGs to the timeline as “captions” for a user. This is dumb for so many reasons: upload/download, using external credit-based transcription services instead of Adobe’s, it’s a destructive process, not editable in timeline, font issues, and many places where process can fail (see Firecut, Cutback, and Brevidy for examples of this process).
To that end:
Per-word timing from the Text tab. Let us read the per-word timing and export to JSON with word-level timestamps from Transcript and Captions tabs.
Start, stop, and edit preferences for transcription/translation processes
Read/Write for every Caption Track & Track Item - including if it is “disabled” aka invisible
Convert Captions to Graphics
Export SRT from timeline
I am likely forgetting other concrete feature requests, but this would go a long way for caption development and is the primary goal.
The secondary issue is a holistic overhaul of all text animation in Premiere Pro. That’s a problem for a different team
Bonus:
Here is a video by “PremiereGal” going thru the pros and cons of 4 caption plugins. I suggest watching this to see the limitations and challenges with the various existing plugins: https://www.youtube.com/watch?v=fPITmB4aMaI
Popping in to say thank you to the Pr team for adding “Export transcript (json)…” and “Import Static Transcript” options to the Text panel!
This targets my #1 request for per-word timing.
I know ExtendScript API updates are on hold in favor of UXP. But having these options in ExtendScript API before the long-tail UXP rollout would be a massive, instant upgrade for caption plugins.
Sorry, I am a newbie here. But I wished to know how can I open and use this JSON files? I tried to use it to automate a script based on text-editing, but couldn’t find a way to read the JSON code exported from premiere. It seems to be a proprietary JSON code. Sorry if it is a dumb question (I’m sure it is!).