Skip to main content
Participating Frequently
December 21, 2021
Question

Transcript-based Lip Sync

  • December 21, 2021
  • 2 replies
  • 675 views

This is a long way up from After Effects rigging and trying to do puppets there. A lot of hard work has gone into this and you guys keep improving all the time.

I can give you details on this issue and the files but first, does Windows 10 have a problem importing the SRT files to use with the "compute lip sync take from audio and transcript"? SRT files do not show for import.

 

Tried pasting a short text script into the "transcript" box I thought maybe my audio was too muddy becuase of being in a song - error message below - so I recorded the same in a "spoken" word track (not singing) with no backgroud music. The audio take was only about 18 seconds long. I got the same error message below.

 

Error message: [Compute Lip Sync failed: Check that the transcript matches the audio.]

 

Any assistance is much appreciated.

 

The "compute lip sync take from scene audio works fairly well.

This topic has been closed for replies.

2 replies

Participating Frequently
December 29, 2021

I managed to do a work-around becuase, as noted the music track was to muddy. I managed to do the whole song with a separate track speaking the words. Most of the "compute lip sync take from audio and transcript" did work and was pretty good.

TheOriginalGC
Community Expert
Community Expert
December 29, 2021

It's pretty easy to confuse CH with things like audio filters and music. Whenever possible, start with a "clean" audio track before the bells and whistles are added and then replace the track after you're happy with the lip-sync.

Participating Frequently
December 21, 2021

The SRT was my mistake. It does show for import. The second issue of [Compute Lip Sync failed: Check that the transcript matches the audio.] Still is an issue.