SongFormer achieves state-of-the-art strict boundary detection and functional label accuracy in music structure analysis by fusing SSL representations and using learned source embeddings on a new 14k-song corpus and expert benchmark.
Evaluation Metrics We evaluate the performance of our proposed SongFormer using the following metrics: (1)HR.5F: The F-measure of boundary hit rate within 0.5 seconds
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
SongFormer: Scaling Music Structure Analysis with Heterogeneous Supervision
SongFormer achieves state-of-the-art strict boundary detection and functional label accuracy in music structure analysis by fusing SSL representations and using learned source embeddings on a new 14k-song corpus and expert benchmark.