بحث

Show HN: Sparrow-1 – Audio-native model for human-level turn-taking without ASR