OpenAI had a lot of eyes when it released ChatGPT-4o to the world earlier this week, and while some of its features are still rolling out, so far it certainly feels

GPT-4o ('O' stands for 'Omni') has made a big change to the chatbot and has almost real emotions behind it. Add a natural-sounding voice to your phone. And now another AI company has demonstrated how to synthesize digital humans using the voice mode of the GPT-4o model.

X, VICTOR Riparbelli, co-founder of the AI video engine Synthesia, said, "The GPT-4o voice mode is really impressive."

"I made a face with my latest avatar model, @synthesiaIO EXPRESS-1. When empathy is important, whether it's healthcare, coaching or education, a friendly face really makes a difference," he added, pointing to the importance of video calling platforms like Zoom over traditional voice-only calls.

We previously covered Synthesia, which produces a kind of avatar for users who are a bit anxious.

In the Riparbelli example, Synthesia produces a woman in a red shirt, and the lip sync is pretty spot-on to what chatgpt-4o's voice mode is saying.

When the demo goes around the room, the video is cut at some point, so it's hard to say if the AI avatar is playing around with the room looking around, but when explaining to the model that the demo is about showing off its abilities, it surprises and does a neat sort of "head tilt."

If you want to check out GPT-4o, you're rolling it out now. Here's how to get access — although OpenAI is ready to wait if it has not yet put you on the list.