Reproduce “One Prompt, Many Sounds: Modeling Listener Variability in LLM-Based Equalization” 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
Ioannis Stylianou, Jon Francombe, Pablo Martinez-Nuevo, Sven Ewan Shepstone, Zheng-Hua Tan
The authors frame language-guided equalization as predicting a distribution of plausible settings, rather than one correct setting, because listeners can interpret a request differently. They collect 11 settings per prompt–audio pair and evaluate text-conditioned models by comparing predicted distributions with listener choices. Tested in-context-learning methods outperform random guessing on both metrics; this measures agreement with the participant pool, not listening satisfaction.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.