Please add the new Kyutai Moshi Voice Model

Launched today, this open-source voice model has a response time of 200ms, and the voices sound as good if not better than the Open GPT-4o voice model that still hasn't been released.

It would be great if you could add this model to the list of supported AI models in Synthflow. This will significantly increase demand for voice assistants as the voices are indistinguishable from human agents.