CSME 2026 Abstracts

Area 1 - Computer Supported Music Education
Full Papers Short Papers

Area 1 - Computer Supported Music Education

Full Papers
Paper Nr: 7
Title:

Piano Practice Support System with Visual Feedback and Evaluation Protocols for Learning Experience Based on Subjective and Objective Metrics
Authors:

Ikumi Osaki, Takako Fujioka, Elizabeth Schumann and Akira Maezawa
Abstract: Many computer-supported music education systems use visual feedback on performance or learning progress as a core feature, making it essential to assess learning effects and learner experience together. We propose an integrated evaluation protocol for such systems and apply it in a case study of a browser-based visual feedback tool for piano practice. The system plots individual note events in a rectangular area (x: time, y: MIDI velocity), allowing learners to inspect temporal and dynamic performance patterns. We investigate: (1) which learner profiles benefit most; (2) whether using feedback is associated with changes in MIDI features (calculated by onset timing, velocity, duration, articulation); and (3) how feedback relates to perceived workload, affect, and usefulness. We conducted a pilot study with amateur pianists of varying skill levels. The perceived value of visual feedback depended on skill level. Objective benefits appeared mainly in note-accuracy–related metrics, while temporal measures showed limited improvement, aligning with earlier work. Subjective measures did not show reliable between-condition differences; interviews suggested that perceived usefulness varied by skill level, indicating that the system’s value may depend on learner expertise and the performance aspect being targeted.
Download
Paper Nr: 9
Title:

MuSA Voice: Evaluating Deep Learning-Based Vocal Technique Feedback for Self-Regulated, Reflective Singing Practice
Authors:

Suvi Häärä, Isabelle Oktay and Rafael Ramirez-Melendez
Abstract: This paper presents MuSA Voice, to our knowledge, the first open web application for multidimensional reflective vocal performance analysis with integrated deep learning (DL) models for Vocal Technique Classification (VTC), and explores its educational potential in a pilot user study. The platform allows users to compare their vocal technique performance against a reference audio, focusing on pitch modulation (straight vs. vibrato) and vocal tone (belting vs. breathiness), with visual feedback, playback, and zooming tools to facilitate post-performance reflection. To assess the validity in the praxis of the implementation, a single-session pilot study with 19 subjects was conducted. In open-ended responses, MuSA Voice was described as engaging and informative, though model inconsistencies under varying recording conditions affected overall trust. Perceived usefulness ratings of the tool were consistently high across all feedback conditions (median = 4/6), with high perceived clarity of the feedback (median = 5/6). Objective performance analysis suggests that the tool is more effective for pitch modulation than vocal tone, possibly due to inconsistent breathy tone classification. These findings offer preliminary insight into the potential of DL-based VTC feedback for independent singing practice and highlight the need for improved model robustness and longitudinal studies.
Download
Paper Nr: 12
Title:

Mumabot: Weaving Music, Mathematics and Educational Robotics in ECEC Settings
Authors:

Maddalena Cocco, Tecla Morettin, Anna Mortara and Luisa Zecca
Abstract: This contribution discusses a pilot course within the “INTEC - Developing an innovative in-service training model for ECEC staff” project (Erasmus+ KA2). The Mumabot (MUsic, MAthematics and educational roBOTics) training course presented an innovative methodology, based on a transdisciplinary and multimodal approach, focusing on the intertwining of the three disciplines through the use of Ozobot, the Rhythm Box and Lego bricks as mediators. The study aims to investigate the effect of the pilot course on the 39 trainees that were involved in the Italian and international versions. Our research questions are: Is this course effective in giving ECEC professionals the ability to plan Mumabot activities? What effect does it have on their ideas about children’s learning in STEAM disciplines? Theoretical and empirical enquiries into STEAM disciplines highlighted the need for training and support for ECEC professionals, and the possibilities offered by a multimodal and transdisciplinary approach. Pre- and post-course questionnaires containing both multiple.choice and open-ended questions were administered to the participants. Thematic analysis was used to analyse the written responses. The findings suggest that while the course was effective in changing participants' perspectives on STEAM activities in ECEC, it was only partially effective in supporting professionals in planning Mumabot activities.
Download
Paper Nr: 13
Title:

A Two-Case Study on Extending Conventional Music Practice Using Mobile Applications
Authors:

Matevž Pesek, Jaka Kužner, Emir Hodžić, Klara Žnideršič and Matija Marolt
Abstract: This article presents a two-case study exploring how mobile applications can extend conventional music practice through playful and interactive digital approaches, addressing two complementary aspects of music education: creative production and rhythmic skill acquisition. The first case describes a mobile application for playful music creation, integrating a sequencer and piano roll into a simple, responsive interface that encourages experimentation and musical “doodling.” The second case features a mobile application for learning polyrhythms, using a poly-star visualization to represent complex rhythmic ratios, where users tap in synchrony with animated beats and receive real-time accuracy feedback. Evaluations using the User Experience Questionnaires and the Technology Acceptance Model (TAM) indicated positive learning potential and identified areas for improvement, such as scoring algorithms and device optimization. Results showed exceptionally high stimulation scores for the sequencer and a highly positive pragmatic quality rating for the polyrhythm tool. Both cases demonstrate that intuitive visualizations and playful interaction can complement traditional music training. Future work includes expanded functionality, collaborative features, dynamic rhythm control, comparative evaluation against existing tools, longitudinal studies in formal teaching settings, and further integration into existing or new game apps and platforms.
Download
Paper Nr: 14
Title:

Music and Mathematics in Technology-Enhanced STEAM Learning Pathways for Lower Secondary Education
Authors:

Alice Barana, Jefferson Curtaz, Veronica Ferrarese and Marina Marchisio Conte
Abstract: This study investigates the design, implementation, and preliminary evaluation of two interdisciplinary learning paths integrating music, mathematics, and digital technologies within a STEAM framework for lower secondary education (Grades 7-8). Grounded in the structural analogies shared by the two disciplines, including pattern recognition, proportional reasoning, and symbolic representation, the intervention was developed using the ADDIE instructional design model, informed by teacher focus groups, and organised around a blended theory-practice-digital methodology. A pilot study was conducted at a lower secondary school in Aosta, involving two treated and two control classes; data were collected through pre- and post-questionnaires on motivation, engagement, and disciplinary knowledge, complemented by platform-based formative assessment and classroom observation. The study produced two structured, institutionally compliant, and replicable learning pathways for Grades 7 and 8, designed to foster meaningful connections between musical and mathematical learning through technology-mediated instruction. Although constrained by sample size, intervention duration, and reliance on self-reported measures, the work provides evidence-informed foundations for STEAM-oriented interdisciplinary curriculum design and establishes a basis for further experimental validation and large-scale implementation.
Download
Short Papers
Paper Nr: 5
Title:

Practical Integration and Critical Perspectives of AI-Powered Applications in K-9 Music Education: Tools, Methods, and Classroom Activities
Authors:

Vladimír Novotný
Abstract: This paper presents a comprehensive framework for integrating AI-powered and web-based applications into music lessons for primary (grades 1–5) and lower secondary (grades 6–9) levels. Utilizing classroom-tested methodologies developed in collaboration with the "AI for Children" (AI dětem) initiative in the Czech Republic, the study demonstrates how tools such as Chrome Music Lab, Suno AI, and Incredibox develop key musical competencies including rhythm, melody, and genre sensitivity. These tools are implemented through the constructivist E-U-R (Evocation-Understanding-Reflection) pedagogical framework. This research addresses two key questions: (1) How can AI tools be effectively integrated into music education while maintaining pedagogical integrity? (2) What are the socio-cultural implications of AI-mediated music learning? Beyond a practical guide for practitioners, this research critically examines the "musical machine habitus," analyzing how cultural dispositions encoded into algorithms may foster Eurocentrism and symbolic violence within digital music systems. By synthesizing a practical instructional guide with a socio-technical critique of cultural homogenization and cognitive offloading, this paper offers a "wise synthesis" for harnessing AI's benefits while preserving authentic musical self-expression and cultural diversity.
Download
Paper Nr: 6
Title:

Interpretable Repertoire Search for Computer Supported Aural Skills Teaching: A Searchable SIA Based Digital Companion
Authors:

Peter G. Fielding
Abstract: Aural skills teaching depends heavily on repertoire selection, yet many large print and heritage song collections remain difficult to search in pedagogically meaningful ways. Teachers may need melodies with limited pitch collections for early solfège, modal examples for comparative listening, or repertory with controlled chromatic complexity, but such features are rarely indexed in printed sources. This paper proposes a computer-supported response to that problem: a searchable, rights-aware digital companion built around tonic-centered successive interval arrays (SIAs). SIAs provide an interpretable representation of pitch space by encoding the cyclic step pattern among the pitch classes present in a melody after tonic normalization. To move beyond a purely conceptual proposal, the paper draws on an already annotated Nova Scotian print corpus of 1,948 songs from 27 publications (Fielding, 2014). Across that corpus, seven-note collections are most common (884; 45.4%), followed by six-note collections (603; 31.0%), with smaller and larger collections occurring less frequently. Query demonstrations show that the metadata can retrieve limited-pitch songs, modal repertory, and melodies with chromatic inflection in ways directly relevant to classroom planning. The paper argues that searchable SIA-based metadata can support curriculum design, culturally grounded repertoire discovery, and computational thinking in music education while keeping printed sources central.
Download
Paper Nr: 8
Title:

Surface Haptics for Music Education: Evaluating TanvasTouch as an Interactive Learning Interface
Authors:

Cristina Greco, Luca Andrea Ludovico and Cristina Rottondi
Abstract: Touchscreen-based music education tools typically rely on audio and visual feedback, despite the inherently sensorimotor nature of musical learning. Recent advances in surface haptics enable programmable tactile feedback on flat displays, opening new possibilities for multisensory interaction. This paper explores the use of TanvasTouch technology in music education, focusing on design patterns, pedagogical applications, and accessibility. We developed interactive activities using tactile textures to represent musical shapes and conducted an exploratory study with primary school students (ages 6–11). Participants engaged in guided touch tasks to recognize simple musical symbols and instrument shapes, allowing us to assess recognition, engagement, and usability. Results show that surface haptics can support exploratory learning experiences, enhancing traditional audio-visual approaches. However, blind shape recognition was largely unsuccessful, indicating that the technology cannot yet function as a standalone modality for visually impaired users. Instead, TanvasTouch is better suited as a complementary multisensory tool in inclusive educational settings.
Download