yogendra-yatnalkar
commented
2 months ago Hi @Linkfufu glad you liked it. Sorry for the late reply.
Which part are you interested in running ?
-
You can generate a bunch of images using text-2-image model of your choice.
-
Later, we can generate the white-board animations using the code: draw-whiteboard-animations.py
- The animation quality will be improved if we tag objects within the image. For tagging, I used a tool called as: AnyLabelling.
- But this step is optional, and the white-board animations can be made without the JSON file (containing segmentation masks) as well.
-
The text I am generating is using googles free text-2-speech API.
-
Lastly, I am concatenating everything: Audio and Video using FFMPEG.
Please let me know if you need any further details any of the above part or any other information, will be happy to help.
Hi, sir, thanks for your work on this amazing project! Could you please give a explaination on how to run this project? Many thanks!