Evaluating High-Resolution Piano Sustain Pedal Depth Estimation with Musically Informed Metrics

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Zhang, Hanwen, Fang, Kun, Wang, Ziyu, Fujinaga, Ichiro
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911417139986432
author Zhang, Hanwen
Fang, Kun
Wang, Ziyu
Fujinaga, Ichiro
author_facet Zhang, Hanwen
Fang, Kun
Wang, Ziyu
Fujinaga, Ichiro
contents Evaluation for continuous piano pedal depth estimation tasks remains incomplete when relying only on conventional frame-level metrics, which overlook musically important features such as direction-change boundaries and pedal curve contours. To provide more interpretable and musically meaningful insights, we propose an evaluation framework that augments standard frame-level metrics with an action-level assessment measuring direction and timing using segments of press/hold/release states and a gesture-level analysis that evaluates contour similarity of each press-release cycle. We apply this framework to compare an audio-only baseline with two variants: one incorporating symbolic information from MIDI, and another trained in a binary-valued setting, all within a unified architecture. Results show that the MIDI-informed model significantly outperforms the others at action and gesture levels, despite modest frame-level gains. These findings demonstrate that our framework captures musically relevant improvements indiscernible by traditional metrics, offering a more practical and effective approach to evaluating pedal depth estimation models.
format Preprint
id arxiv_https___arxiv_org_abs_2510_03750
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Evaluating High-Resolution Piano Sustain Pedal Depth Estimation with Musically Informed Metrics
Zhang, Hanwen
Fang, Kun
Wang, Ziyu
Fujinaga, Ichiro
Information Retrieval
Sound
Audio and Speech Processing
Evaluation for continuous piano pedal depth estimation tasks remains incomplete when relying only on conventional frame-level metrics, which overlook musically important features such as direction-change boundaries and pedal curve contours. To provide more interpretable and musically meaningful insights, we propose an evaluation framework that augments standard frame-level metrics with an action-level assessment measuring direction and timing using segments of press/hold/release states and a gesture-level analysis that evaluates contour similarity of each press-release cycle. We apply this framework to compare an audio-only baseline with two variants: one incorporating symbolic information from MIDI, and another trained in a binary-valued setting, all within a unified architecture. Results show that the MIDI-informed model significantly outperforms the others at action and gesture levels, despite modest frame-level gains. These findings demonstrate that our framework captures musically relevant improvements indiscernible by traditional metrics, offering a more practical and effective approach to evaluating pedal depth estimation models.
title Evaluating High-Resolution Piano Sustain Pedal Depth Estimation with Musically Informed Metrics
topic Information Retrieval
Sound
Audio and Speech Processing
url https://arxiv.org/abs/2510.03750