Show HN: Sparrow-1 – Audio-native model for human-level turn-taking without ASR (tavus.io) 123 pts|1 month ago|49 comments
