Subtitling based on provided text instead of AI analysis ?

Salut !

Is it possible to use Resolve's subtitle generation feature by feeding it a clean & correct text (with no mistakes and proper punctuation etc...), and just ask it to align that text with the sound, instead of Resolve generating the text himself (with all the mistakes it induces) ? I have all the dialog written down properly, so I would rather exploit that version !

I think YouTube can do that, give it a text file and it will align it with the audio to generate the subtitles. Hopefully Resolve can at least do what a mainstream platform like YouTube can do, and do it even better, especially since in a professional setting, scripts are usually written in the first place, so this feature would be even more elementary than creating quick & dirty captions for TokTok videos, which is currently pretty much the realistic extent of its capability..

If it can detect the text from audio, surely it has to be able to also align a given text to the audio. Right ? ... right ?

Thanks

The feature you describe doesn't exist in Resolve, but it wouldn't be too difficult to implement. I wonder if there are third party subtitle editors that do this and then export an SRT.

The feature you describe doesn't exist in Resolve, but it wouldn't be too difficult to implement. I wonder if there are third party subtitle editors that do this and then export an SRT.

hi Igor,
Thank you for jumping in !
Yes I've been searching for 3rd party like crazy with no luck so far. I even posted on Reddit and everyone immediately considered I was a moron for possibly wanting something like this. I'm a bit surprised since in the professional world a lot of the times, we do have the Script and much rather use that than spend 3 hours fixing the AI !
Good luck to all of us

You can upload a subtitle file to YouTube, but it is more than just the correct text, it also includes time stamps. Otherwise, how would Google know which words to put where? AI generates the subtitle with timing information because it knows at what point it translated the audio into text and can place that in the file.