Create Your Own Voice Avatar | Talking Face Animation with A Single Image and Audio Clip |SadTalker

% 4

I am demoing a free open-source tool for audio2face, a type of Generative AI. It allows you to create your own speakingface animation simply from a single static image and some audio sample! That is, bringing image to live with GenAI technology. SadTalker stands for Stylized Audio-Driven Single Image Talking Face Animation.
How to download presentations, notebooks & source codes of the video?
- Click 'Subscribe' to access and stay updated on the latest AI tutorials
- Click my Google Drive share link to download:
drive.google.com/drive/folders/1bMkRUztSkFKlvZ1TMUd5j-IwivCKDl31?usp=drive_link
- Download all AI notebooks and PPT resources
drive.google.com/drive/folders/1ez7EKwF9doLA3cVFjZzpBS03xCXkvkYp

Наука

Опубликовано:

16 июн 2023

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист

Посмотреть позже

Комментарии : 3

@msalama Год назад

thanks, how to make it faster so we can use it in real time applications

@aitutorials2681 Год назад

Next step I am going to try GPU machine to see if big difference.

@aitutorials2681 Год назад

@msalama, Tested on a GPU server, the result: GPU: 26s audio clip -> 66s processing time: ratio 2.53 CPU: 5s audio clip -> 30+mins processing time: ratio 1800/5=360 GPU:CPU speed boost = 360/2.53 = 142 times faster!