analiza aiqbench
This paper addresses a core bottleneck in affective computing by releasing a rigorously annotated dataset of 2,496 tracks alongside DAMER, a novel framework that fuses Mel spectrograms and cochleagrams to improve emotion recognition robustness. The work demonstrates significant gains over existing methods on standard benchmarks like PMEmo and 1000songs while offering open-source resources for reproducible research in audio AI.
Memo2496: Expert-Annotated Dataset and Dual-view Adaptive Framework for Music Emotion Recognition
arXiv:2512.13998v3 Announce Type: replace-cross Abstract: Music Emotion Recognition (MER) is constrained by limited expert annotations and the need to establish robustness across heterogeneous corpora. Memo2496 supplies a reproducible dataset of 2,496 instrumental tracks with continuous valence-arousal labels from 30 certified music specialists, supported by interface familiarisation and duplicate-track intra-annotator calibration in a normalised circular domain. We also introduce the Dual-view Adaptive Music Emotion Recogniser (DAMER), a general framework evaluated on Memo2496 and two external datasets. DAMER integrates Dual-Stream Attention Fusion (DSAF) for token-level bidirectional interaction between Mel spectrograms and cochleagrams, Progressive Confidence Labelling (PCL) for curriculum-based pseudo-labels using temperature scheduling and Jensen-Shannon divergence, and Style-Anchored Memory Learning (SAML), whose labelled contrastive queue regularises same-emotion embeddings across acoustically varied samples. The primary evaluation follows the binary MER protocol used on PMEmo and 1000songs, while a supplementary continuous regression study demonstrates direct use of Memo2496 segment-level valence and arousal scores. Experiments on Memo2496, 1000songs, and PMEmo show that DAMER achieves the highest arousal accuracy among compared methods on Memo2496 and 1000songs and the highest valence accuracy on PMEmo, while remaining competitive for PMEmo arousal. Ablations and diagnostics validate each module. The dataset and source code are publicly available.
Abstract:Music Emotion Recognition (MER) is constrained by limited expert annotations and the need to establish robustness across heterogeneous corpora. Memo2496 supplies a reproducible dataset of 2,496 instrumental tracks with continuous valence-arousal labels from 30 certified music specialists, supported by interface familiarisation and duplicate-track intra-annotator calibration in a normalised circular domain. We also introduce the Dual-view Adaptive Music Emotion Recogniser (DAMER), a general framework evaluated on Memo2496 and two external datasets. DAMER integrates Dual-Stream Attention Fusion (DSAF) for token-level bidirectional interaction between Mel spectrograms and cochleagrams, Progressive Confidence Labelling (PCL) for curriculum-based pseudo-labels using temperature scheduling and Jensen-Shannon divergence, and Style-Anchored Memory Learning (SAML), whose labelled contrastive queue regularises same-emotion embeddings across acoustically varied samples. The primary evaluation follows the binary MER protocol used on PMEmo and 1000songs, while a supplementary continuous regression study demonstrates direct use of Memo2496 segment-level valence and arousal scores. Experiments on Memo2496, 1000songs, and PMEmo show that DAMER achieves the highest arousal accuracy among compared methods on Memo2496 and 1000songs and the highest valence accuracy on PMEmo, while remaining competitive for PMEmo arousal. Ablations and diagnostics validate each module. The dataset and source code are publicly available.
Q. Li, C. L. P. Chen and T. Zhang, "Memo2496: Expert-Annotated Dataset and Dual-View Adaptive Framework for Music Emotion Recognition," in IEEE Transactions on Affective Computing
Sound (cs.SD); Artificial Intelligence (cs.AI); Multimedia (cs.MM)
arXiv:2512.13998 [cs.SD]
(or arXiv:2512.13998v3 [cs.SD] for this version)
https://doi.org/10.1109/TAFFC.2026.3715195 DOI(s) linking to related resources
Text integral sub licența deschisă a publicației · 238 cuvinte
Subiecte conexe
Build a Physical AI model factory with NVIDIA Cosmos 3 on SageMaker HyperPod
AWS · 4 sept. 2026
[AINews] Reality Checks on AI News (Yegge shuts down Gas Town, Databricks’ +60% Astra cost)
Latent Space · 17 sept. 2026
FUSEP: A Multi-Center Benchmark for Diverse Tasks in Early Pregnancy Fetal Ultrasound Screening
arXiv · 6 aug. 2026
Introducing CARE-X: Towards Clinically Useful Radiology VLMs with Auxiliary Supervision, Reward-Aligned Learning, and Tool-Augmented Measurement
Microsoft Research · 11 aug. 2026