1. Upload videos(or images)

Deepfakes are videos in which the subject is face-swapped using machine-learning algorithms. You can easily create Deepfakes video on Deepfakes web β.
Everything works in the cloud. All you need to do is upload videos.
Only you can access your learning data and videos.
You can reuse your trained model. This allows you to improve the quality of the results.
1. Upload videos(or images)
2. Wait until completion
3. Done!
4. Reuse your model
$2 / 1 hour of usage rights
Estimated time required:
Learn from video/images and face swap → About 4 hours
Face swap by reusing a trained model only → About 30 min
Deep learning requires a lot of GPU processing power, which is quite expensive on the cloud.
Thank you for understanding our prices.
* The quality of the result depends on learning data. It is not guaranteed.
* For use within the range stipulated by the terms of use and laws and regulations
Uchimura's Best Video Award 2019/10/22
Deepfakes Web was used and introduced in the TV Tokyo program “Uchimura's Best Video Award”.