OpenAI showcased Sora's cutting-edge multi modal live demos including a remarkable showcase where they created a fully narrated video using Sora and their voice engine.
In this demo, OpenAI takes us beyond mere conversation. Using the Mac app from ChatGPT, they've combined audio conversations with their Assistant API's text-based conversation, enriched with vision support. But what's truly fascinating is their venture into multimodality with Sora, a diffusion model capable of generating videos from prompts.
#openai seamlessly creates a script to narrate the visuals generated by #sora They slice a few frames, feed them to #gpt4o in real-time, and voila! A story unfolds before our eyes, all thanks to the synergy of vision capabilities and AI storytelling.