Show HN: Caramel: a speech-to-speech conversation model trained only on audio

This source publishes only a summary in its feed, so the full article lives at the outlet. The link above and below goes to the original.