A project made to leverage AI tools to generate a mostly automoated video geneation pipeline.
This uses selenium to acess CHATGPT and uses prompts to generate scrips from a selected webpage content, Chat GPT to get prompts to image geneation and then using those prompts to get base image content.
This is used with a ComfyUI pipeline to locally turn the image into an animation, a TTS libray is used to read the scipt and a streamlined process is used to manually select the best animations to use for the video geneation.
Video is assembled using FFMPEG.