Batch Editing Speaker Names in Transcripts

[dbeaty2]

I have a documentary with a voice over artist that edits his takes as individual wav clips. So I have a folder with over 100 VO clips. I can transcribe these to do Text based Editing. But I can't find a way to change ALL the hundreds of clips of the same voice Speaker Name to the narrator. 

It seems this would be metadata? I can't find Speaker in any of the metadata lists unfortunately. So I guess its saved outside the actual clip within the transcipt.  I'd like to be able to change the speaker of all clips I select as a batch function. In some cases this would cause problems when there are multiple speakers, so it would have limited use to clips that are all the same person. But still a very usuful feature. 

[Stan Jones]

Upvoted, but I'd look for a workaround.

You are correct; it is not metadata. I believe the Speaker info is embedded with the text and timecode info in Base64 encoded data in the project file. No way to access that.

Also, Speaker information is not imported to the transcript using the "import corrected transcript" method.

How are you using the wav files? Are they laid down one after the other on a timeline? I'd consider exporting them to get one transcript file.

Stan

[Peter Sinter]

Is there any solution or workaround? I have hundreds of clips for a documentary and would like premiere to detect and name the same speaker in all clips. Is there really no way and I would have to do this on a clip basis? I usually put all clips in a sequene to preview them but then I am not able to rename the speakers as premiere only allows to edit speaker names in the clips itself. This way the speaker name function is useless.. Or does it maybe work with a compound clip?

[Stan Jones]

@Peter Sinter,

See below for my thoughts about a workaround.

The developers are adding the ability to bulk delete speakers (only in the Beta so far), and they may have other ideas in store:

https://community.adobe.com/t5/premiere-pro-beta-discussions/text-based-editing-update-bulk-delete-s...

So it is a good time to suggest options and to describe your workflow. What is your use case? You only mention one speaker.

This is a test of a workaround. A single speaker across a whole project will be easier, so I tested 3 speakers - Jim, Mary, and Bob. I have many clips of Jim and Mary. I have only one of Bob. I put the clips of Jim and Mary into a sequence called "Clips". (They could be in whatever order on the timeline you wish.) I create a static transcription of that sequence. (If I add a sequence marker at every clip, I'll have a clip marker at that point when the sequence is nested.) When nested into another sequence, this will behave as a "clip" capable of using text-based editing.You could also place Jim in one sequence and Mary in another and nest those separately.

I create a master sequence ("Editing") and add the Bob clip. (Before or after adding, I transcribe Bob as source media.) I nest Clips into Editing - as a nest, not as individual clips.

With no editing of Speaker names, there will be 2 Speaker 1's: Jim is Speaker 1 in Clips and Mary is Speaker 2. Bob is Speaker 1 in Editing - so in Editing, we have 2 Speaker 1's.

To edit speaker names,  you must be in the view for the type of transcript you have. Even though the Clips transcript was created as a static transcript, we are using it as a clip transcript. So to edit the speakers there, in the Project Panel, right-click "Clips" and "Open in Source Monitor." Switch to the Text Panel -> Transcript Tab. You can edit the Speaker names there. So Speaker 1 is now Jim and Speaker 2 is Mary. Do the same with the Bob clip, so that Speaker 1 is now Bob.

Now in Editing, there are 3 speakers each with a different name.

I can work in the Editing transcript to select sections and lift/extract, paste, etc.

@TeresaDemel FYI. Can you share any more about the road map for Speakers?

Stan

[Peter Sinter]

Thanks @Stan Jones !

You are right: I have a lot of different speakers, not only only one.

I was hoping I could just rename those speakers once in the full transcript of the sequence of all clips and it gets transferred across all clips. Its just for convenience, for example when I am using the transcript as a script for my video later on the speakers are already there. I didnt think about the fact that premiere isnt able to track speakers across multiple clips in the first place, so the problem is not renaming at all. Thanks for your workaround, but I will avoid nesting clips at all times - they can always cause issues and you will notice mostly in the end of the project when meeting a deadline.. So all in all I can only hope for an update from premiere to track speakers across all clips in the same bin. AND I would really hope some time it will be posssible to edit the whole script for the final movie in the text window - so I can write down a text intended for the overvoice as well while keeping a dynamic script where I can use text based editing.. but maybe thats a thing for the next five years. 😉

[Stan Jones]

Thanks for the response with your workflow issues. Here's hoping for continuing progress in this area.

Stan