Copy link to clipboard
Copied
Earlier this year we debuted Text-Based Editing in Premiere Pro which makes editing video as simple as copying and pasting text. We’ve been excited to see all the ways you’re utilizing this workflow to speed up your rough cut – and we’ve been listening to all the feedback. Now in Premiere Pro (beta) we’re bringing you some of the top feature requests so far: filler words, bulk actions (like remove all pauses), and multichannel audio support.
(If you’re new to Text-Based Editing, check out our HelpX page or explore the workflow with Valentina Vee’s complete beginner’s guide to Premiere Pro.)
Filler Words
Text-Based Editing now allows you to detect “uh” and “umm” filler words, so your transcript is more accurate. Just like with pauses, you can click on a filler word and delete it from your sequence transcript. Because we serve an international audience, Filler words are language agnostic so that they work in all 18 languages supported in Speech to Text.
When you transcribe a sequence, you will now see a localized version of “[filler]” to identify these filler words. If you’ve previously transcribed a clip or sequence, you’ll need to re-transcribe to add filler words to it.
Bulk Actions (including Remove All Pauses)
Now you can get rid of all those filler words, pauses, or any other words you don’t want to keep in your sequence transcript in a single click. This is an easy way to clean up everything from long silences to brief pauses in an instant.
Open the transcribed sequence you want to trim. In the Transcript panel, click on the filter icon and choose filler or pause.
Every instance of filler words or pauses will become highlighted in the transcript. Click Delete and select Extract (to ripple edit and close gaps) or Lift (to delete and leave gaps). You can Delete all at once or delete one by one.
If you only want to get rid of long pauses or fillers, open up the overflow menu->Transcribe view options. You can change the length of pauses and fillers. The transcript will update in real time.
You can also search for text and delete it, as well as replacing instances of text displayed on the source or sequence. For example, you can find and replace every instance of a proper noun which may have been transcribed incorrectly.
Transcript View Options
You also have more control over what you see in the transcript. Click on the three dots at the top of the Transcript window and select Transcript view options. You can add visual indicators for low-confidence words, adjust pause length, and display or hide un-transcribed sources, filler, pauses, and speakers.
You also have more control over search settings with the ability to narrow a search to whole words only or match capitalization.
Multichannel Audio Support
Some audio files have different microphones recorded on separate channels. With multichannel audio support in Text-Based Editing, you can choose to re-transcribe an audio file using a specific channel or a mix of all channels. This feature works with audio files in a multichannel mono format.
After importing a multichannel mono audio file, right click on the file in the Project panel and choose Transcribe or Re-transcribe. You will now see a channel selection drop down so you can choose if you’d like to transcribe a mix or specific channel.
These great new Text-Based Editing updates are now available in Premiere Pro (beta). Let us know in the comments below how they’re working for you. In particular, we want to know how filler words are performing in non-English languages. We also want to know what we can add to make your workflows feel more smooth and speedy, so keep the requests coming!
Copy link to clipboard
Copied
This is perfect! just tried it on some old ~20 min interviews and it worked great.
Two things I would add. First, is the option to add a crossfade to the cuts of 2-5 frames so it smooths out the transitions automatically. Currently, you can select all and +add crossfade, if you have your duration already set to that duration but would be nice to have it in the delete options as well.
Second, it would be useful if the bulk delete also acted as a "bulk cut" or "bulk copy" and you were able to copy/paste searched phrases and words into the clipboard. Right now if you wan't to copy a bunch phrases or takes of a sentence you have to do it one at a time.
For example, if you are interviewing a CEO and he is on a teleprompter, but he flubbed the last line of the best take, and you need to find all the other takes where he said that same thing and show it to client, If you could bulk copy/paste those segments into a new timeline that save a lot of time, even over the already fast text search function.
Copy link to clipboard
Copied
Along these same lines It would be nice to filter and bulk delete the interviewer and add their question as a text layer, caption, or marker to the timeline.
How this would work:
- Filter button
- Select speaker (interviewer)
- Bulk delete speaker
- Add option to take transcribed speaker you are deleting and add it as a caption or text graphic over the footage to easily reference the deleted question. Maybe add an option for markers as well but I think text layer makes the most sense for a lot of editors.
Potential issues:
-Many times the interviewer isn't recorded clearly, maybe have a way to filter out low volume chatter.
Copy link to clipboard
Copied
Any plans to increase the pause threshold to minutes or making it all pauses no matter how small or big?
Copy link to clipboard
Copied
Can you mass delete all pauses (nor ripple delete) so that just the audio is deleted, but that the mp4 remains so there are no 'blank spaces' in the video line?
Copy link to clipboard
Copied
Have you tried locking the video track and running the remove pauses with a Lift operation instead of Extract (aka ripple delete)?
Copy link to clipboard
Copied
Hey. It doesn't work with russian language. Buttons are grey and i can't push on them. I know its beta but maybe someone knows how to solve it? Also i tried to transcribe speech in englisn, it works, but when i cut pauses it cut some words too. Because Premiere pro can't recognize russian words in english:)
Copy link to clipboard
Copied
Copy link to clipboard
Copied
Feedback
Copy link to clipboard
Copied
Is there a best practice for using Text-Based Editing with a Multicam sequence?
Background:
I have a 3-hour interview recorded with 2 cameras and a lav-mic.
My goal is to produce both short clips and longer cuts from this interview.
Queries:
1. Should I first transcribe all videos from both cameras (approximately 10 separate transcriptions due to 4-5 videos per camera)? Or, would it be more efficient to first set up the Multicam sequence and then transcribe just that?
2. How does the text-based cutting process work?
Can I simply cut from the main sequence and paste into new sequences for each desired video output? Or is there an optimal method to edit within the Multicam sequence itself?
3. When should I start switching between cameras? Since each switch results in a cut, I want to be strategic about this and not start until I have a clear plan.
I'd greatly appreciate any insights or recommendations. Thank you!
Copy link to clipboard
Copied
Amazing! Bring all this to Audition too and I save a lot of money on external services :😆
Copy link to clipboard
Copied
For whatever reason, the AI doesn't appear to picking up any filler words. Has anybody else encountered this?
Copy link to clipboard
Copied
This feature has been removed in the latest update.
No idea why. Kevin Monahan said it should still be in the Beta version.
Copy link to clipboard
Copied
Would love to use this, but I don't have the option to select filler words, nor is the ripple delete option working. Any info on why?
Copy link to clipboard
Copied
This feature has been removed in the latest update.
No idea why. Kevin Monahan said it should still be in the Beta version.
Copy link to clipboard
Copied
The "filler word" option under the Transcript view is nowhere to be found. Anyone who also experience the same issue?
Copy link to clipboard
Copied
What version of Premiere are you on?