Adaptive Turn-Taking for Real-time Multi-Party Voice Agents
For developers of multi-party voice agents, this work addresses the challenge of dynamic floor competition with a role-conditioned approach that yields substantial performance gains.
ModeratorLM improves turn-taking in multi-party voice agents by conditioning on an assigned role, achieving over 40% higher precision and 70% higher recall in turn-taking, with fewer false-positive interruptions.
Turn-taking in multi-party spoken conversations remains a fundamental challenge for voice-based agents, particularly under dynamic floor competition and varying user expectations. We propose ModeratorLM, a role-playing voice agent that conditions turn-taking behavior on an explicitly assigned role in multi-party settings. The system is built on a speech large language model operating in chunk-wise streaming manner. We further introduce a reasoning-augmented variant that incorporates chain-of-thought reasoning over conversational context and the assigned role. We construct RolePlayConv, a large-scale synthetic dataset of spoken multi-party conversations with diverse assistant roles. Experiments on real-world meeting data and RolePlayConv show improved turn-taking precision by over 40% and recall by more than 70%, while substantially reducing false-positive interruptions compared to non-role-conditioned baselines.