作者位次 #6/11ACL 2026Poster· Natural Language Processing
Towards Fine-Grained and Multi-Granular Contrastive Language-Speech Pre-training
Yifan Yang · Bing Han · Hui Wang · Wei Wang · Ziyang Ma · Long Zhou · Zengrui Jin · Guanrou Yang · Tianrui Wang · Xu Tan · Xie Chen
摘要
Modeling fine-grained speaking styles remains challenging for language-speech representation pre-training, as existing speech-text models are typically trained with coarse captions or task-specific supervision, and scalable fine-grained style annotations are unavailable. We present FCaps, a large...