Full-DuplexBench

Emerging

3papers using it

2026first seen

Full-DuplexBench is a benchmark dataset used to evaluate the performance of full-duplex speech-to-speech dialogue systems, focusing on turn-taking and conversational intelligence.

🔎 Find this dataset

Papers using Full-DuplexBench (3)

PersonaPlex: Voice and Role Control for Full Duplex Conversational Speech Models2026

Liberating LLM Capabilities in Full-Duplex Speech Models2026

DuplexCascade: Full-Duplex Speech-to-Speech Dialogue with VAD-Free Cascaded ASR-LLM-TTS Pipeline and Micro-Turn Optimization2026