Style embeddings computed from CLAP and a sound matching model align with human timbre similarity ratings better than plain embeddings, MFCC, and other pre-trained representations.
We choose to report triplet agreement instead, as it is a more intuitive metric to interpret than the other two
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SD 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Assessing the Alignment of Audio Representations with Timbre Similarity Ratings
Style embeddings computed from CLAP and a sound matching model align with human timbre similarity ratings better than plain embeddings, MFCC, and other pre-trained representations.