Videos

RealTalk: We Recreated Joe Rogan's Voice Using Artificial Intelligence



Dessa

NEW: The latest phase of our work raising public awareness with deepfakes it out, this time featuring a hyper-realistic deepfake of Joe Rogan that combines audio and video. Check it out here: https://youtu.be/i7QNUZWS6VE

Check out The New York Times’ coverage of this work in their television show The Weekly: https://www.nytimes.com/2019/11/22/the-weekly/deepfake-joe-rogan.html

This video and the model that it demonstrates was created by three Machine Learning Engineers at Dessa (Hashiam Kadhim, Rayhane Mama, Joseph Palermo).

Learn more:
Part 1 of our blog on RealTalk discusses the importance of raising awareness about speech synthesis and its societal implications:
https://medium.com/@dessa_/real-talk-speech-synthesis-5dd0897eef7f

Part 2 of our blog takes a look under the hood of RealTalk and provides a technical overview of how it works:
https://medium.com/dessa-news/realtalk-how-it-works-94c1afda62f0

The RealTalk system was developed on Foundations Atlas, Dessa’s product for rapidly training, testing and deploying cutting-edge machine learning models. Learn more about Atlas and Dessa on our website: atlas.dessa.com , or check us out on Product Hunt!
https://www.producthunt.com/posts/foundations-atlas

Please note that this project does not suggest that we endorse the views and opinions of Joe Rogan.

Source

Similar Posts

37 thoughts on “RealTalk: We Recreated Joe Rogan's Voice Using Artificial Intelligence
  1. The emotions are variable in the same sentence its very obvious that its not a human voice. Each word seems to have a different feeling and its not like hes telling the same story. Obvious compilation of words.

    This is very simple software, unlike the Jordan Peterson Deepfake

  2. I think you just took different words he has said and matched them together. No proof you actually used AI for any of this (but you get more clicks claiming you did)

  3. Soon the recorded confessions of criminal activities will not be admissable in court and combined with deep fake imagery the video won't be either. Maybe it already should be. The future sure is crazy

  4. I don't get it how can the computer simulate him taking the breath before letting out a sentence, if anything it would just say the sentence with no pause to take a breath.

Comments are closed.

WP2Social Auto Publish Powered By : XYZScripts.com