作者位次 #11/14ACL 2026Findings· Findings of ACL
Z ip V oice-Dialog: Non-Autoregressive Spoken Dialogue Generation with Flow Matching
Han Zhu · Wei Kang · Liyong Guo · Zengwei Yao · Fangjun Kuang · Weiji Zhuang · Zhaoqing Li · Zhifeng Han · Dong Zhang · Xin Zhang · Xingchen Song · Lingxuan Ye · Long Lin · Daniel Povey
摘要
Generating spoken dialogue is inherently more complex than monologue text-to-speech (TTS), as it demands both realistic turn-taking and the maintenance of distinct speaker timbres. While existing autoregressive (AR) models have made progress, they often suffer from high inference latency and stab...