Accurately parse tiktok videos and analyse them into text using gpt4v

Open source address:https://github.com/disingn/cliptalk
Introduction: Using Google gemini-pro-vision and gemini-pro or GPT4-vision and GPT4 to parse Shake and tiktok video content , analyse it into text content , in addition to the accompanying Shake and tiktok watermarking interface
I am relatively new, big brother do not spray, to provide you with an idea.