DuplexCascade: 全双工语音到语音对话的无VAD级联ASR-LLM-TTS流水线与微轮询优化
DuplexCascade: Full-Duplex Speech-to-Speech Dialogue with VAD-Free Cascaded ASR-LLM-TTS Pipeline and Micro-Turn Optimization
AI总结：
DuplexCascade通过无VAD级联流水线和微轮询优化，实现全双工语音对话的高效交互与强对话智能。
摘要：
具有级联ASR-LLM-TTS模块的语音对话系统保留了强大的LLM智能，但VAD分割往往迫使半双工轮次并导致脆弱的控制。另一方面，无VAD的端到端模型支持全双工交互但难以维持对话智能。在本文中，我们提出了DuplexCascade，一种无VAD的级联流式流水线，用于全双工语音到语音对话。我们的核心思想是将传统的按语句划分的长轮次转换为按片段划分的微轮次交互，从而在保持强大文本LLM优势的同时实现快速双向交流。为了可靠地协调轮次交替和响应时间，我们引入了一组对话专用控制标记，以在流式约束下引导LLM的行为。在Full-DuplexBench和VoiceBench上，DuplexCascade在全双工轮次交替和开放源语音到语音对话系统中的对话智能方面达到了最先进的水平。
英文摘要：
Spoken dialog systems with cascaded ASR-LLM-TTS modules retain strong LLM intelligence, but VAD segmentation often forces half-duplex turns and brittle control. On the other hand, VAD-free end-to-end model support full-duplex interaction but is hard to maintain conversational intelligence. In this paper, we present DuplexCascade, a VAD-free cascaded streaming pipeline for full-duplex speech-to-speech dialogue. Our key idea is to convert conventional utterance-wise long turns into chunk-wise micro-turn interactions, enabling rapid bidirectional exchange while preserving the strengths of a capable text LLM. To reliably coordinate turn-taking and response timing, we introduce a set of conversational special control tokens that steer the LLM's behavior under streaming constraints. On Full-DuplexBench and VoiceBench, DuplexCascade delivers state-of-the-art full-duplex turn-taking and strong conversational intelligence among open-source speech-to-speech dialogue systems.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
