Reproduce “CF-AVSS: A Coarse-to-Fine Audio-Visual Speech Separation Framework with Target-Speaker Refinement and Post-Fusion for Real-World AVSE Challenge” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Longjie Luo, Zhongshu Hou, Yuepeng Li, Gang Yang, Shidong Shang, Fan Zhou
CF-AVSS separates two overlapping speakers in noisy, reverberant recordings using two video-aligned coarse estimates, then refines them with the original mixture and fuses coarse and refined outputs. On the challenge test set, it had the lowest character error rate on both tracks, tied first on Track 1 and ranked second on Track 2 by OVRL, an average of metric ranks.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.