• Global community
    • Language:
      • Deutsch
      • English
      • Español
      • Français
      • Português
  • 日本語コミュニティ
    Dedicated community for Japanese speakers
  • 한국 커뮤니티
    Dedicated community for Korean speakers
Exit
1

Text Based Editing Speaker Removal

Community Beginner ,
May 21, 2024 May 21, 2024

Copy link to clipboard

Copied

I'm trying out Text Based Editing in Premiere on the Beta branch and have run into an issue.

 

I transcribed the whole sequence and then made it a static transcription so I could assign Speakers correctly. But now, there's no way to just go in and delete a single Speaker like I want to since the button is greyed out. And when I click the button to redo it for text based editing, my whole app crashes (repeatedly). I think it may have something to do with the fact that I cut out all of the silence and umms, so now there's like a thousand edits and it is treating each as its own little file instead of the original 4 files it was at first. This also means I can't really do any Source editing since there's so many files (at least, I don't think?).

 

Any ideas?

Views

300

Translate

Translate

Report

Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines

correct answers 1 Correct answer

Adobe Employee , Jun 03, 2024 Jun 03, 2024

Hi Trevor,

thanks a lot for breaking down your workflow, I'll try to answer all your questions.

 

To avoid some confusion, I want to make sure the difference between a static and a dynamic transcript is clear. When you want work with text-based editing, you are creating dynamic transcripts of your source clips, meaning each source clip has it's own transcript attached, and wherever you put that clip in your timeline, its transcript will follow. It's dynamic because changes you make in the timeli

...

Votes

Translate

Translate
Community Expert ,
May 22, 2024 May 22, 2024

Copy link to clipboard

Copied

Moved to beta forum.

provide feedback.png

Votes

Translate

Translate

Report

Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
Adobe Employee ,
May 23, 2024 May 23, 2024

Copy link to clipboard

Copied

Hello @TrevorTheEditor,

 

thank you for flagging this issue! I have a few follow-up questions regarding the crash you're seeing:

 

- Did you transcribe all your source clips before you started working on your sequence?

- You want to switch from a static sequence to creating a text-based edting transcript - when exactly did the crash occur? Right aways after clicking on the "Transcribe" button?

- When the crash occured, did you see the Crash Report window opening up? If yes, please put in your user email and send the crash report so I can search for the crash log to get more information what exactly is crashing (I can't reproduce the crash on my end)

- Are you working on Mac or Windows?

 

A few tips that might also help your workflow:

- You can change and edit the speaker name in your source transcripts, the change will automatically update in your sequence transcript (this is only the case when working with test-based editing transcripts, not static transcripts)

- In static transcripts it's not possible to remove the speakers from the transcript/timeline because the transcript is - static. Any changes you make in the timeline will not update in the transcript. We recommend using static transcripts at the end of your workflow once you have a picture lock. The static transcription will transcribe only the audio in your timeline and will not attach transcripts to yor source files.

 

Best, 

Kerstin

Votes

Translate

Translate

Report

Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
Community Beginner ,
May 24, 2024 May 24, 2024

Copy link to clipboard

Copied

thanks for sharing your knoledge. 

But is there any way to delete a entire speaker with the Text-based Editing tool?

Votes

Translate

Translate

Report

Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
Community Beginner ,
May 24, 2024 May 24, 2024

Copy link to clipboard

Copied

I think the developers should have link somehow the text info to the metadata so we can easily edit by speaker inside the Premiere. Or at least make something to show what speaker is in the 'captions' pannel. that way we can edit faster and make more money to pay Creative Cloud 😄

Votes

Translate

Translate

Report

Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
Adobe Employee ,
May 27, 2024 May 27, 2024

Copy link to clipboard

Copied

Hi @Felipe5C95 

 

Yes, you can delete an entire speaker in a text-based editing sequence. Just add your transcribed source clip to a sequence and then filter for the speaker name you want to delete (Transcript tab -> click on filter icon next to search bar in top left -> filter for speakers -> click "Delete" button -> select "lift" or "Extract" and then delete single instances or all instances of that speaker with "Delete/Delete all"). This will delete all parts with that speaker in your sequence.

You can't search for a speaker in the Captions panel though. Captions only show the spoken dialogue but not the speaker info. So it's best to stay in the Transcript tab if you want to delete speakers in a sequence.

 

Best,

Kerstin

Votes

Translate

Translate

Report

Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
Community Beginner ,
May 28, 2024 May 28, 2024

Copy link to clipboard

Copied

Okay so I think one of my issues was that I needed to click on the TIMELINE and not the Source in order to get the Delete button to appear.

Secondly, I believe that the program simply doesn't like it if you remove all silence and nonsense words before removing a Speaker. It looks like if I remove a Speaker FIRST and then do the others, it likes that much better and doesn't have the crash situation when having to re-transcribe all the little bits to then get the Speaker assignments.

 

As a side note, how long is transcription supposed to take exactly? I have about an hour and a half of content and it gives me an ETA of about 5 hours, is that normal?

Votes

Translate

Translate

Report

Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
Community Beginner ,
May 28, 2024 May 28, 2024

Copy link to clipboard

Copied

Not normal. Here it lasts about 5-10min for a timeline about 1:30 hour.

I use a Mac Mini m1

Votes

Translate

Translate

Report

Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
Community Beginner ,
May 28, 2024 May 28, 2024

Copy link to clipboard

Copied

Thanks Kerstin and Adobe community! 

Votes

Translate

Translate

Report

Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
Community Beginner ,
May 28, 2024 May 28, 2024

Copy link to clipboard

Copied

Thanks for the follow up. Answers as best as I can:

-I transcribed the whole sequence, not individual items. I keep seeing tutorials referencing transcribing through the Source Monitor, but that sounds...backwards? Can I not just drag all my items into my timeline and hit Transcribe? My objective was to be able to sync my audio and video files and then generate a transcription so it would then let me use text based editing to effect the synced collection. Do I have to sync them together first, export, then re-import so they are one file each to make this work right?

 

-I tried to make the system work without static transcription, but it wouldn't let me assign individual speakers until I made it static. It just listed everything as Unknown.

 

-No crash window popped up. Premiere just closed without warning or error at all.

-I am working on Win11

 

Okay, so what it sounds like from your direction is that I should be making almost all of my changes in the Source window rather than the timeline itself, yes? And it should still update directly in the timeline once changes are made? And I should be avoiding static transcription if I am trying to do text based editing, correct?

Votes

Translate

Translate

Report

Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
Community Beginner ,
May 28, 2024 May 28, 2024

Copy link to clipboard

Copied

if you want to transcribe a sequence you should synk first, for sure. 

Votes

Translate

Translate

Report

Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
Adobe Employee ,
Jun 03, 2024 Jun 03, 2024

Copy link to clipboard

Copied

Hi Trevor,

thanks a lot for breaking down your workflow, I'll try to answer all your questions.

 

To avoid some confusion, I want to make sure the difference between a static and a dynamic transcript is clear. When you want work with text-based editing, you are creating dynamic transcripts of your source clips, meaning each source clip has it's own transcript attached, and wherever you put that clip in your timeline, its transcript will follow. It's dynamic because changes you make in the timeline or the Text panel  (e.g. changing clip order, deleting parts) will update automatically in both instances.

A static transcription only transcribes the audio of your sequence, not the individual clips. Changes that you make on the sequence after transcribing are not synced with the transcript.

 

  • Re transcription duration: it depends on how many source clips you have in your timeline, and how long these source clips are. For example, if your timeline is only 5 minutes long but contains parts of 20 clips, all 20 clips will be transcribed fully. A static transcription would be faster at that point because it only looks at the audio in the sequence, not the source clips attached to it.
  • You can only delete speakers in a transcript when you are in a dynamic sequence transcript. It's not possible in a source transcript because you can't delete parts of a source clip within the source clip itself. However, when you are in a source clip transcript, you can edit/change the speaker name (change it from Unknown to the correct name).
  • You should be able to edit speaker names once you see the "..." before speaker name or Unknown. Click on the dots, then click on "Edit speaker name".. Also, if you enable "Speaker labelling" in your transcription settings, different speakers will be detected and segmented automatically.
  • I suggest you sync your a/v files first and then transcribe the synced clip (but you can skip the "export->import again" step, just transcribe the synced media).
  • I still don't fully understand in which situation you are seeing the crash. If it is repdoducible, could you make a screen recording (including the steps you did right before the crash) and share it with me?

 

Thanks,

Kerstin

Votes

Translate

Translate

Report

Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
Community Beginner ,
Jun 03, 2024 Jun 03, 2024

Copy link to clipboard

Copied

LATEST

Okay so it looks like the order in which things are done is incredibly important as doing them out of order causes a crash. The correct order seems to be to remove specific Speakers FIRST before doing any other removals. Additionally, based on the notes I recieved here, WHERE specifically you are clicking when trying to remove content is a huge deal. Sometimes you have to double click on a clip in the timeline to put it in Source before editing, othertimes you have to click on the timeline first.

I would say overall, the workflow is pretty clunky as it has a lot of invisible rules that don't allow much deviation or notification you are doing something wrong, but as of right now, the issue has been solved and I appriciate all the help 😃

Votes

Translate

Translate

Report

Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
Resources