Copy link to clipboard
Copied
My program can create a static transcription with no problem, but I can't seem to make a Text-Based Editing Transcript on any clip I import or drop into a sequence.
The text panel states that I do not have any Dialogue Audio from my clip.
Am I missing something?
Hey, thanks for the quick response. I just tried the updated version of the Beta and received the same results. I took my original audio files into a sequence on their own and it didn't recognize it as dialogue. In essential sound I marked it as dialogue and still no luck.
Copy link to clipboard
Copied
I have gotten the message when I think I should not, but I cannot see a pattern.
The most consistent option is to open the clip in the Source Monitor before adding to a sequence. And from the Source Monitor, go to the Text Panel, Transcript tab.
You will get this message if you mute the audio or if there is, in fact, no dialogue.
Stan
Copy link to clipboard
Copied
Hi there, I am one of the engineers working on Text Based Editing. Not sure I fully understand your workflow but if I understand it correctly -- this should work. Could you make sure you are on the latest beta and share a screen recording of the issue you're describing with me?
Copy link to clipboard
Copied
Hi there!
I am having an issue with transcription and never have before. I was using 24.0 when the issue began, I have 2 audio tracks and 3 speakers, but only 1 of the speakers' audios will actually transcribe.
My workflow is normally: Windows> Text> Transcript> Transcribe sequence and then create captions from there. This only transcribed the roughly 36 seconds as you can see in the 3rd image. I have tried updating to 24.0.3 which didnt work, reverting to an old beta version, uninstalling PrPro all together, reinstalling an older version (which wouldnt open the project as it was a newer version), and then ultimately reinstalling 24.0.3. The clips are all marked as dialogue in essential sound as shown and when I select "transcribe Audio track 1" it shows no dialogue.
I even tried exporting the video, bringing it back into premiere and attempting to have it transcribe the mixed down audio and it still only recognized the same ~25 sec.
Any help would be great!
Copy link to clipboard
Copied
happy to take a look at this. Any chance you could try this in the current beta version (we made some fixes in that area so there's a chance your issue is already fixed. You can install Beta in parallel to the normal version).
If that doesn't help - would it be possible for you to send me the project you're using + the audio assets. I don't need the video assets but would be great to get the audio.
Thanks & Greetings, Nico
Copy link to clipboard
Copied
Hey, thanks for the quick response. I just tried the updated version of the Beta and received the same results. I took my original audio files into a sequence on their own and it didn't recognize it as dialogue. In essential sound I marked it as dialogue and still no luck.
Copy link to clipboard
Copied
Copy link to clipboard
Copied
Thank you for sending me the audio files. I have to admit - this one was a head scratcher. There was no real reason this sequence shouldn't transcribe. Finally found the culprit in the raw audio files. Seems when recording these someone set up the stereo mics in a way that they produce a 180 degree phase shift. In other words - the left and right track cancel each other out.
You don't hear that in the sequence or when exporting as stereo - but when rendering the audio as mono (which is required for transcription) the left and the right channel sum up to almost exactly zero (+ a bit or residual noise).
So in this case the fix is "relatively" easy. Just open the raw audio files in Audition, select either the left or the right channel, use Effects->Invert and save the file. This should make transcription work again.
Copy link to clipboard
Copied
Wow, this is crazy - thank you for looking into this! Its a bit odd thought because all of the audio files were recorded the exact same way on the same equipment, even the same input. We will be sure to keep this in mind in the future as a fix if we run into this again.
Really appreciate your help!