Reproduce “Perceptual Differences and Preferences in Speech Enhancement” 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
Simon Dahl Jepsen, Jesper Rindom Jensen
The authors evaluate speech-enhancement systems by measuring whether listeners can distinguish outputs and whether they prefer one, treating these as separate outcomes. In a crowdsourced study, listeners assessed four systems on 45 four-second utterances from the DNS test set using triangle tests and paired preference tests. All tested pair averages exceeded chance discrimination, ranging from 0.56 to 0.83; incomplete pairings limit direct comparisons.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.