Unit-Based Agent for Semi-Cascaded Full-Duplex Dialogue Systems
This work addresses the challenge of creating more natural human-computer dialogue systems, though it appears incremental as it builds on existing multimodal large language models and auxiliary modules.
The paper tackles the problem of enabling natural full-duplex voice interaction by decomposing dialogue into minimal conversational units, allowing independent processing and transition prediction, and achieves second place on the test set of the Human-like Spoken Dialogue Systems Challenge.
Full-duplex voice interaction is crucial for natural human computer interaction. We present a framework that decomposes complex dialogue into minimal conversational units, enabling the system to process each unit independently and predict when to transit to the next. This framework is instantiated as a semi-cascaded full-duplex dialogue system built around a multimodal large language model, supported by auxiliary modules such as voice activity detection (VAD) and text-to-speech (TTS) synthesis. The resulting system operates in a train-free, plug-and-play manner. Experiments on the HumDial dataset demonstrate the effectiveness of our framework, which ranks second among all teams on the test set of the Human-like Spoken Dialogue Systems Challenge (Track 2: Full-Duplex Interaction). Code is available at the GitHub repository https://github.com/yu-haoyuan/fd-badcat.