Creative Communities of the World Forums

The peer to peer support community for media production professionals.

Activity Forums Adobe Premiere Pro Speech-to-text dual-system sound problem!

  • Speech-to-text dual-system sound problem!

    Posted by Nate Tam on November 2, 2012 at 12:14 am

    Hey guys,

    So I’m working on a documentary piece, and I’m trying to figure out this whole speech-to-text thing.

    It’s working and and analyzing everything okay, but as soon as I hit the spacebar, the words on screen and the audio dont match.

    What I mean is the timecode that the speech-to-text is telling me is wrong.

    We DID shoot dual-system sound (separate audio and video) and I synched them in premiere. So my question really is, is Premiere analyzing the audio timecode and telling me what words are in which spots, but not incorporating that to the merged clip (meaning the video timecode)?

    The words are simply not lining up with the video. UNLESS i play the audio alone. Then it works like its supposed to.

    Nate Tam
    Freelance Film Editor
    natetamiam.com

    Alex Udell replied 13 years, 9 months ago 2 Members · 5 Replies
  • 5 Replies
  • Alex Udell

    November 2, 2012 at 12:59 pm

    In your dual audio…..your audio TC is different than you video TC?

    Alex Udell
    Editing, Motion Graphics, and Visual FX

  • Nate Tam

    November 2, 2012 at 4:16 pm

    Yes. The video and audio timecode are different but the clips are merged. The timecode of the merged clip follows the video timecode, not the audio, and I think that’s what the problem is…I just don’t know how to fix it.

    Nate Tam
    Freelance Film Editor
    natetamiam.com

  • Alex Udell

    November 2, 2012 at 8:11 pm

    It sounds to me like you need to take your synced clip…

    export it as a new master….

    then use that as the editing and transcription source….

    Alex

  • Nate Tam

    November 2, 2012 at 9:22 pm

    I could do that, but I’m working with RED raws. Transcoding and re-conforming is not something I can do with this much footage as well. Let alone the time it takes to do so.

    Is there no other way besides burning the clips together via exporting to solve this problem?

    Nate Tam
    Freelance Film Editor
    natetamiam.com

  • Alex Udell

    November 4, 2012 at 5:15 pm

    Hey Nate…

    Yep that would be painful….

    The problem, it seems, is that the speech to text is using the Audio source timecode, not the video source timecode…do I have that right?

    Maybe there is a way to re-timecode the audiomedia to match the video timecode….that way the transcript will then match…

    this is only from a quick google: https://www.videotoolshed.com/product/42/qtchange

    but there ought to be something out there that will let you do this…

    Alex Udell
    Editing, Motion Graphics, and Visual FX

We use anonymous cookies to give you the best experience we can.
Our Privacy policy | GDPR Policy