PHALAR: Phasors for Learned Musical Audio Representations

Davide Marincione; Michele Mancusi; Giorgio Strano; Luca Cerovaz; Donato Crisostomi; Roberto Ribuoli; Emanuele Rodolà

cs.SDMay 5, 2026

PHALAR: Phasors for Learned Musical Audio Representations

Authors: Davide Marincione, Michele Mancusi, Giorgio Strano, Luca Cerovaz, Donato Crisostomi, Roberto Ribuoli, Emanuele Rodolà

Organizations: Department of Computer Science, Sapienza University of Rome, Italy · Moises Systems, Inc. · Paradigma, Inc.

Abstract

Stem retrieval, the task of matching missing stems to a given audio submix, is a key challenge currently limited by models that discard temporal information. We introduce PHALAR, a contrastive framework achieving a relative accuracy increase of up to $\approx 70\%$ over the state-of-the-art while requiring $<50\%$ of the parameters and a 7 $\times$ training speedup. By utilizing a Learned Spectral Pooling layer and a complex-valued head, PHALAR enforces pitch-equivariant and phase-equivariant biases. PHALAR establishes new retrieval state-of-the-art across MoisesDB, Slakh, and ChocoChorales, correlating significantly higher with human coherence judgment than semantic baselines. Finally, zero-shot beat tracking and linear chord probing confirm that PHALAR captures robust musical structures beyond the retrieval task.

Explore similar work

FIGMA: Towards FIne-Grained Music retrievAl
Jun 4, 2026Nishit Anand, Ashish Seth, Sreyan Ghosh +2Music Understanding
Frequency-Aware Self-Supervised Music Representation Learning
Jun 24, 2026Yicheng Gu, Junan Zhang, Jerry Li +2Mel-Frequency Cepstral Coefficient Music Understanding
MERIT: Learning Disentangled Music Representations for Audio Similarity
May 26, 2026Abhinaba Roy, Junyi Liang, Dorien HerremansMusic Understanding Audio Latent

Jun 4, 2026 · cs.SDJ/K move · Enter open · S save

FIGMA: Towards FIne-Grained Music retrievAl

Nishit Anand, Ashish Seth, Sreyan Ghosh, Dinesh Manocha+1

University of Maryland, College Park, USA

Retrieving music using natural language descriptions has improved with contrastive audio-text models such as CLAP, but current systems remain limited to coarse semantic queries. When descriptions specify fine-grained musical attributes such as tempo, key, chord progression, or rhythmic structure, existing models often fail to retrieve the correct audio. We show that this limitation stems from the contrastive learning objective itself: despite being trained on long captions, CLAP-based models effectively utilize only the first few tokens, discarding much of the information encoded in detailed prompts. Then, we propose FIGMA (FIne-Grained Music RetrievAl), a multi-view contrastive architecture that addresses this limitation by jointly optimizing global audio-text alignment and frame-level, token-wise alignment. This design enables FIGMA to capture both high-level semantic context and fine-grained musical attributes within a unified representation space. Moreover, we formalize the task of Fine-Grained Music Retrieval and construct Fine-Grained Music Caption dataset (FGMCaps), a large-scale dataset of 380K music-caption pairs for training along with a 10K test set, both annotated with tempo, key, chord progression, beat count, as well as genre and mood. Extensive experiments demonstrate that FIGMA consistently outperforms existing CLAP-based music retrieval models across multiple music retrieval benchmarks, including out-of-domain evaluations, with relative improvements of up to 73.3%.

Music Understanding

Open paper Ask about it PDF

Abstract

Explore similar work

FIGMA: Towards FIne-Grained Music retrievAl

Frequency-Aware Self-Supervised Music Representation Learning

MERIT: Learning Disentangled Music Representations for Audio Similarity