A framework learns to map seed music embeddings to mood-adjusted targets using proxy sampling and a joint objective, outperforming baselines in preserving non-mood attributes on two datasets.
Supervised and unsupervised learning of audio representations for music understanding
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SD 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Controllable Embedding Transformation for Mood-Guided Music Retrieval
A framework learns to map seed music embeddings to mood-adjusted targets using proxy sampling and a joint objective, outperforming baselines in preserving non-mood attributes on two datasets.