DuplexAgent-RSI separates live speech from async agent work in voice assistants
jiqizhixin · x · 2026-11-04
CUHK-Shenzhen and Amphion Technology introduce DuplexAgent-RSI, separating continuous conversation from asynchronous search, reasoning, and coding in voice assistants.
- Problem: Hearing an interruption isn't enough—the system must update active tasks, reject stale results, and pick proper speaking windows while staying responsive.
- Mechanism: A collaboration harness tracks task identities, versions, and delivery state; a timed simulator generates hard interactions, and failure diagnosis guides targeted code repairs; paired comparisons and guard checks gate changes while interaction-model weights stay fixed.
- Results: Ten improvement rounds substantially improved collaboration in the VoiceChat experiment.
More from coding & agent
- Karpathy's viral take: engineers' biggest regret is unsaved AI conversations — bfrench · 2026-11-04
- Apple lets apps bundle Siri skill files, devs mock the sloppy docs — msg · 2026-11-04
- YC-backed Simantic adds a simulation layer so AI agents can test firmware at software speed — ycombinator · 2026-11-04
- Bittensor Subnet 33 Launches Skill Store: Tested Skills Aim to Make Open Models Do Frontier Work — markjeffrey · 2026-11-04
- Building a 24-hour AI newsroom on Microsoft Copilot Harness: workflow beats prompts — WirelessLife · 2026-11-04
- Polyphonic to give AI agents their own Mac via open-source terminal-mac — RileyRalmuto · 2026-11-04