Search papers, labs, and topics across Lattice.
University of Padova, Padova, Italy
4
0
6
2
Taste-sound correspondences in AI-generated music vary significantly across cultures, with response-style biases obscuring deeper perceptual differences.
Achieving high-quality semantic audio generation on low-resource devices without sacrificing performance or complexity is now a reality with \textit{aria}.
Predicting taste from audio embeddings not only surpasses human consensus but also redefines the benchmarks for music retrieval systems.
Expertly curated data beats sheer data quantity: a small, musicologically-informed LilyPond dataset outperforms a 150x larger MIDI corpus for music understanding tasks.
