Reproduce “Identifying birdsong syllables without labelled data” 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
Mélisande Teng, Julien Boussard, David Rolnick, Hugo Larochelle
The authors develop an unsupervised method that discovers recurring birdsong syllable templates and matches them to recordings without labeled syllable training data. It detects events in spectrograms, clusters events, and uses template matching to recover syllable sequences. Across four Bengalese finches, multi-individual event-detection precision was 0.82 and syllable-class precision 0.91, but recall was 0.57; evidence is limited to clean recordings.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.