Reproduce “Reverberant Speech Conditioning for Material-Aware Acoustic Parameter Mapping” 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
Yi Wu, Olha Townsend, Qianqian Fang, Magdalena Fuentes, You Zhang
The authors estimate room-acoustic maps from a floor plan and reverberant speech recorded at one talker–microphone pair, using speech-derived conditioning with a U-Net predictor. On MRAS, Rec-RIR reduces mean proportional T30 error from 31.5% to 19.9%, compared with 17.5% using a reference response. Gains mainly improve map levels, with little improvement in spatial variation, and real-room transfer is mixed.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.