Voice notes in, spoken answers out — dictate audio that becomes user messages (transcribe), have the agent read replies aloud (speak), and leave walk-away narration on long headless runs. Local-first: plain audio files under ~/.dsh/voice/.
Give a DeepSeek Harness agent a voice it owns — offer_call rings the human (接听/拒接/稍后); local TTS via CrispASR + Qwen3-TTS CustomVoice, 9 speakers, 2 Chinese dialects. Local-first, offline-capable.