Skip to main content
Participant
April 19, 2023

Text-Based-Editing Doesn't work with multiple audio channels

  • April 19, 2023
  • 29 replies
  • 11892 views

I frequently record videos with more than one audio source. anywhere from 2 to 4 channels. I've found out if the microphone audio is not on the primary 1st track, it doesnt transcribe anything and says its silent.

Essentially Text Based Editing Workspace only transcribes videos with 1 audio track. So i have to edit the clip and reencode it with only one layer of audio

There needs to be an option just like subtitles to select the track to encode

29 replies

Known Participant
October 13, 2023

@Stan Jones 

 

This is how my audio channels look like:

 

 

The single speaker is on clip 2.

 

The transcription options show no channel selections

 

 

Stan Jones
Community Expert
Community Expert
October 13, 2023

@vsbmeza,

 

I suspect for most of us, the files that work vs the ones that don't will be related to this comment in the guide: "This feature works with audio files in a multichannel mono format." In other words, not every multi-channel audio will work.

 

Look at your file in Modify -> Audio Channels. This is what a 4-channel mono looks like:

 

 

If the audio is NOT multichannel, the option to pick channels will not show in the Transcribe dialogue.

 

If one of the mono tracks actually has multiple speakers in that one track, it will transcribe, but you will obviously not get what you want.

 

This is still being actively worked on in the latest Beta versions.

 

Discussions there are a bit complicated, because bugs are expected and come and go.

 

Stan

 

 

 

Known Participant
October 13, 2023

@Stan Jones while that might have been the intention, it still doesn't work with video files that have multiple audio channels in PR 24.0.0 Build 58.

 

Stan Jones
Community Expert
Community Expert
October 13, 2023

@vsbmeza,

 

> 6 months and a proper release later, this is still an issue. Can we expect this to work soon?

 

Transcribing multichannel clips is included in the new PR 24 release. See "Multichannel audio support" on the revised Text-Editing guide:

https://helpx.adobe.com/premiere-pro/using/text-based-editing.html

 

Stan

 

Known Participant
October 12, 2023

6 months and a proper release later, this is still an issue. Can we expect this to work soon?

Stan Jones
Community Expert
Community Expert
August 30, 2023

@Craig31673466nf06,

 

True. I suspect it is a matter of the many directions possible now that the text-based editing has rolled out.

 

Upvote this feature request. Add a comment with any thoughts you have about how you'd like it to work.

https://community.adobe.com/t5/premiere-pro-ideas/audio-track-selection-for-source-media-transcription/idi-p/13828019

 

Stan

 

Participant
August 29, 2023

If descript can make this work, I don't see why its a problem for Adobe.

Rocman
Participant
August 2, 2023

Premiere 24 transcription accuracy went down the tubes.

The Accuracy has gone way down. Words, Phrases and whole sentences are totally missing, Great Disappointment!

Your're suggested work around doesn't seem to help much unless I'm totally not understanding what you are suggesting. Help! 

slovill
Inspiring
July 1, 2023

I would like to thank Adobe for offering such a wonderfully limited feature without telling us about the limitation. I just pissed away two hours of my day trying to make something work that Adobe never intended to work but sold anyway. Almost as befuddling as limiting this feature to channel 1 is Adobe's decision to not inform the user base of this limitation. Editors around the globe are now frustrated and wasting time trying to use a tool with an unknown limitation that presents itself as a bug. Does Adobe not have any folks who do actual video editing working there?

Stan Jones
Community Expert
Community Expert
June 30, 2023

@Mauronic,

 

I tried to find a way to map audio channels so source media transcription would treat the correct one as "track 1" audio. But it appears to look at the actual file and not any way that PR interprets the audio.

 

Drag the file from the Project Panel to the new icon so it creates a matching sequence.  Then mute all the audio except one at a time, and determine which track has the dialogue you want. Create a static transcript using that track.

 

If only one channel of a track has the dialogue, and the mix on that track does not work, it get difficult.

 

See the workarounds discussed in the feature request:

https://community.adobe.com/t5/premiere-pro-ideas/audio-track-selection-for-source-media-transcription/idi-p/13828019

 

Stan