Adobe Creative Cloud

Digital Video & Audio

November 21, 2014 /

Update on Speech to Text functionality in video tools

When it was originally implemented 6 years ago, the Speech to Text feature was based on state of the art technology that we anticipated would improve at a faster rate than it did.  Today, after many years of work, we believe, and users confirm, that the Speech to Text implementation does not provide the experience expected by Premiere Pro users. Based on this, and feedback from customers, we have decided to remove the feature from the next update to Premiere Pro, Prelude and AME in order to focus our resources on other features.

The metadata generated by Speech to Text in existing projects will remain(it is embedded) and will persist moving forward to future versions and will be usable and accessible. Though the “content (speech) analysis” feature to create new speech analysis text will no longer be available in new versions, you will be able to view the text for any clips that have previously been analyzed and had speech text generated in previous versions (8.1 and earlier).

If you still need to use the feature, you can install CC 2013 Release (or earlier) and run a Speech to Text analysis on your media before bringing it into the current version of Premiere Pro CC.  All metadata created by Speech to Text analysis will remain associated to your media, will be usable and accessible.

If you rely on this feature, make sure to read the release notes before you update so you can prepare for the workaround. Stay tuned for more information on the upcoming updates to Premiere Pro CC, Prelude CC and Adobe Media Encoder CC and detailed documentation on the work around options.

Join the discussion

  • By Michal - 12:36 PM on November 22, 2014  

    I kinda hoped it will get better with every version of Premiere:/ I guess the technology isn`t that advanced yet.. but thanks for being honest!;-)

  • By Pierre Louis Beranek - 7:08 PM on November 22, 2014  

    Hopefully this frees up resources to fix things that are long overdue, such as:

    1. Project window named ‘Project’. Period. Just like other windows are just called the name of the window. No need to explain why yet again. FR already sent.
    2. Fix bug of focus shifting away from Timeline when toggling in/out of full screen view. Once again, no explanation needed. This is an obvious problem any person who understands editing can see, and constantly affects people using Premiere on a laptop.
    3. Fix bug of focus shifting between Source and Program monitor when both windows are stacked.
    4. Fix problem of effect presets all too easily getting renamed when click/dragging and following step with any key press. Come on guys, this is long overdue and incredibly stupid!
    5. etc, etc, etc.

    Should I go on?

    Thanks for all your hard work. I just wish all of that hard work was on things we actually need and use. At least letting go of a feature very few people need/use is a step in the right direction.

    • By Meagan Keane - 6:08 AM on December 10, 2014  

      Hi Peierre – thanks for your feedback! Hopefully you’re also submitting formal feature requests/bug fixes to adobe.com/go/wish – our product team reads through every single submission.

  • By Jeff Handy - 5:34 AM on November 23, 2014  

    Will this affect the captioning tools?

    • By Meagan Keane - 6:07 AM on December 10, 2014  

      Hi Jeff – No, this should have any effect on the captioning tools.

  • By Andy Field - 11:46 AM on November 24, 2014  

    Disappointed you are pulling this from future releases – we do A LOT of documentary and news work and if someone speaks standard unaccented English it works quite well – it’s an enormous time saver …also removing it removes the AVID like Script Sync function….hope you reconsider as this is a very valuable tool

  • By Anthony Burokas - 1:52 PM on November 27, 2014  

    Both Mac and PC have excellent speech recognition capability. Why not hook into that for better transcriptions? Not only do you get multilingual support, but FAR better results, and less engineering workload on your part?

  • By Angelo - 9:47 AM on December 5, 2014  

    Has Visual Speech Editor (Shown at the most recent MAX) been cancelled as well? Seems silly the Audition team would be working on, essentially, the same feature that Premiere is removing. If so, at least collaborate between teams.

    • By Meagan Keane - 6:06 AM on December 10, 2014  

      Visual Speech Editor was part of MAX sneaks – new technology we are exploring but has not been (and may never) be implemented depending on progress of our research. But its great to hear you’re excited about it!