The seamless transition demo is fantastic. The translated voice is passable for my own native voice. It would be incredible when we can achieve this in real-time.
We released inference code and weights, you can check our github here: https://github.com/kyutai-labs/hibiki