August 15, 2024
I've spent a while today talking to Carter, a video avatar from the team at @heytavus.
Really impressive — fast responses, excellent video and audio generation, nicely leverages the strengths of today's SOTA LLMs and vision models.
You can build your own video avatars and real-time video conversation features on top of Tavus's models and APIs. The docs are great and it's easy to get started.
I'm a little bit biased here. Partly because I'm just generally fascinated by the evolution of real-time, multimodal AI, so I'm rooting for anybody building in this space. Also, when you're talking to Carter, the low-level audio and video networking parts of the conversation are routing through @trydaily's WebRTC cloud. And the Tavus team is contributing to the Open Source @pipecat_ai framework for real-time AI that I also contribute to!
Check out what Tavus is doing. Talking to Carter — or building your own avatar — is a window into the multimodal AI near future.
The new @heytavus avatars launched on Product Hunt today, too!
Go check out the conversation there if you're interested in this tech. The team is fielding questions and feature requests.