M. Zanoni, R. Rossi, A. Sironi, P. Belluco, and C. Rottondi, “SCHuBERT: a real-time end-to-end model for piano music emotion recognition,” in Proc. AVARIG 2026: Audio for Virtual and Augmented Reality and Immersive Games, Jun. 2026, pp. 10313:1-10. [Online]. Available: https://aes.org/publications/elibrary-page/?id=23275
Zanoni M, Rossi R, Sironi A, Belluco P, Rottondi C. SCHuBERT: a real-time end-to-end model for piano music emotion recognition. In: AVARIG 2026: Audio for Virtual and Augmented Reality and Immersive Games. Audio Engineering Society; 2026.pp. 10313:1-10. Available from: https://aes.org/publications/elibrary-page/?id=23275
@inproceedings{Zanoni2026_23275,
author = {Zanoni, Massimiliano and Rossi, Riccardo and Sironi, Alice and Belluco, Paulo and Rottondi, Christina},
title = {{SCHuBERT: a real-time end-to-end model for piano music emotion recognition}},
booktitle = {AVARIG 2026: Audio for Virtual and Augmented Reality and Immersive Games},
pages = {10313:1-10},
url = {https://aes.org/publications/elibrary-page/?id=23275},
year = {2026},
month = jun,
publisher = {Audio Engineering Society},
address = {Paris France}
}
TY - CPAPER
TI - SCHuBERT: a real-time end-to-end model for piano music emotion recognition
AU - Zanoni, Massimiliano
AU - Rossi, Riccardo
AU - Sironi, Alice
AU - Belluco, Paulo
AU - Rottondi, Christina
T2 - AVARIG 2026: Audio for Virtual and Augmented Reality and Immersive Games
SP - 1
EP - 10
M1 - Paper 10313
PY - 2026
DA - 2026/06/30
UR - https://aes.org/publications/elibrary-page/?id=23275
PB - Audio Engineering Society
CY - Paris France
LA - en
AB - In this study, we present SCHuBERT, a real-time end-to-end Piano Music Emotion Recognition (PMER) system that operates directly on raw audio and fine-tunes DistilHuBERT for short-window classification on the ValenceArousal plane. Designed for low latency and high responsiveness, the system is particularly well suited for immersive applications such as virtual and augmented reality. Compared with both audio- and symbolic-domain baselines, SCHuBERT achieves strong accuracy in four-quadrant classification as well as in binary arousal and valence tasks, while maintaining low computational overhead for real-time operation.
ER -