Accessible digital musical instruments (ADMIs) are a promising approach to music-making, demonstrating that technology can play a transformative role in promoting inclusivity in musical practice. In this paper, the author's industrial Ph.D. project is presented, a prototype of ADMI, developed together with the company Audio Modeling and the founder of Musica Senza Confini . The prototype was tested in live ensemble performances involving both users with disabilities and professional musicians, yielding encouraging and insightful results regarding its practical applicability and musical integration. This paper contributes to the broader conversation on inclusion in digital music-making and ofers a concrete step toward the development of tools that accommodate diversity without compromise. Accessibility, usability, digital music instruments, accessible digital musical instruments, human-computer CHItaly 2025: Technologies and Methodologies of Human-Computer Interaction in the Third Millenium, Doctoral Consortium, ∗Corresponding author.
In the field of Human-Computer Interaction (HCI), music plays an important role in its form as a
digital instrument, whether it has an interface or not. We can also assert that music plays a crucial
role in the interaction between humans and the computer when this union intentionally generates
sound. Adding the aspects of accessibility and inclusivity, the field leads directly to accessible digital
musical instruments (ADMIs), intended as accessible musical control interfaces used in electronic music,
inclusive music practice and music therapy settings[
CEUR Workshop
ISSN1613-0073 undertaken, detailing the research question, experimental protocol, testing procedures, and outcomes derived from empirical data. Section 5 brings together the findings of the author and outlines the main takeaways from the study.
The development of ADMIs has become a central concern in the intersection of music technology
and HCI. These systems aim to broaden access to musical expression for individuals with physical
or cognitive impairments, and have found applications in inclusive education, music therapy, and
performance contexts[
The project prototype presented in this paper, UniMIDIHub, focus of the author’s research, is a flexible software application developed to enable MIDI communication across multiple platforms. The concept originated from the founders of Musica Senza Confini (a musical initiative aimed at children and adults with psychophysical disabilities, focused on inclusive music making as part of its work on accessible ensemble performance), who was tasked with creating performance strategies for a user with extremely limited mobility, only able to move their fingers, toes, and eyes.
Based on these assumptions, the research question on which the research and subsequent developments were based concerns the possibility that the software project could be aimed at individuals without a specific disability or ability, but rather at everyone, regardless of their ability. In fact, ADMIs dedicated to a specific disability are more common than this approach. To follow this path, the research framework planned to test the software on people with diferent types of disability, one at a time, and then conduct an initial observational study aimed at people with unique and diferent characteristics, together. to understand how the software works, as shown in Figure 1, the user is able to play a song with UniMIDIHub, reproducing musical chords and sequences through the movement of his eyes, fingers and toes. The system leveraged an eye tracker, routinely used by the user for his daily activities, to let her select colored pads on the screen, while micromuscular sensors triggered control actions mapped to notes, chords, short MIDI sequences, or other sounds. Thanks to its configurable design, UniMIDIHub can be adapted to diferent user needs and ofers a broader selection of pieces to play.
The software runs on a central computer, receives input from a variety of devices, and sends MIDI messages to a digital audio work station (DAW) hosted on a separate system. Within the DAW, a multitrack project is loaded and each track can be assigned to a diferent input device, enabling collaborative and inclusive musical performances.
UniMIDIHub is developed in C++ through the JUCE framework and is compatible with both macOS and Windows. Consists of a standalone user interface with configurable colored pads, ranging from 2 to 12 per screen (Figure 2), that can be activated through diferent type of input devices, facilitating accessibility for diverse user needs.
Through the Settings menu, each colored pad can be fully customized with a range of parameters, including: • Input Type: Trigger that activates the sound on mouse hover; Hold that activates on hover and deactivates when the cursor exits the area; Latch, that toggles the sound on hover, deactivates on the next hover; • Generated Action: assignable to a MIDI Note, MIDI Control Change, MIDI Chord, MIDI Sequence, a sample, or a navigation command (e.g., page change); • MIDI settings: channel, note, velocity, release-velocity values, chord properties, and adjustable delay in milliseconds for both MIDI Note-On and Note-Of messages; • Visuals: custom pads color; • Input controls: two main keyboard shortcuts and an optional shortcut to repeat the last triggered note; • Output for the event: MIDI Out, which sends the message to a DAW, to be played, or Sampler, an additional view to set various type of music samples.
In the Setting menu it is also possible to switch between ”Main Pads” (the colored ones) and ”Aux Pads”, which are additional configurable pads that function identically to the main ones but are hidden from the main interface. These pads are triggered exclusively via keyboard shortcuts. In this way, any external device, such as video game controllers, extendable keyboards, accessible devices, etc. that can be mapped with keyboard keys, can “play” UniMIDIHub. For every setup, it is possible to save, load, and reset the configuration, with the opportunity to save numerous setups to load at the time of the performance. There are two possible screen views: the ”Eye Tracker” Mode (see Figure 2) which provides a rectangular area with rounded corners in the center of the screen, as a rest area for the gaze and as a transit area from one pad to another non-adjacent one. The ”Touch Screen” Mode which instead shows the pads in the entire screen and allows easier performance with touch devices. One of the standout features of UniMIDIHub is its integration with the MIDI protocol. As a universal standard for digital music communication, MIDI allows the software to interface with a wide array of tools, including DAWs, plugin hosts, virtual instruments, and external MIDI hardware, enabling maximum workflow flexibility and compatibility. Due to its cross-platform compatibility, configurable interface, and comprehensive MIDI functionality, UniMIDIHub ofers a flexible toolset for music interaction. It supports a range of use cases, including sample triggering, virtual instrument performance, and efect manipulation, thereby facilitating diverse approaches to music production and control. The developments achieved so far have led to the release of a first marketable version on Audio Modeling channels. This step is not considered a point of arrival, but rather a starting point for sales to the public. Research and development will continue even beyond my Ph.D.
Two observational studies, informal uses, and musical performances were held with UniMIDIHub.
The software developments achieved so far have taken into account all the feedback received. The
observational studies aimed at evaluating the usability and accessibility of the software prototype. The
ifrs empirical investigation[
A second observational study was conducted to evaluate UniMIDIHub following the integration of improvements based on feedback from the initial user study. This iteration involved seven participants, all of whom were students enrolled in a Sound and Music Computing program. The study aimed to investigate the following research question: Can students with a background in music technology understand and operate UniMIDIHub independently, and would they propose any additions or modifications to its design? As in the previous study, participants were introduced to the context and purpose of the software but received no prior demonstration or detailed instructions on its use. They were then asked to complete a series of tasks with increasing levels of complexity. The results revealed two primary trends. On one hand, several participants expressed a desire for a more streamlined interface or the inclusion of contextual guidance to enhance usability. On the other hand, despite an initial period of uncertainty, all participants were ultimately able to explore and interact with the software independently. These findings suggest that the application is, in principle, accessible to users with relevant domain knowledge and that its interface can support self-directed exploration, though further refinements may improve the overall user experience. The feedback gathered from this second study has also informed the ongoing development of UniMIDIHub. The current version, incorporating these refinements, has subsequently been tested in live ensemble performances involving both users with disabilities and professional musicians, yielding encouraging and insightful results regarding its practical applicability and musical integration.
In the field of HCI, accessibility remains one of the primary challenges. For software to be considered truly usable and inclusive, it must be accessible to individuals with a wide range of abilities and disabilities. This concern is particularly significant in the context of digital musical instruments, which, through their transition from analog to digital, have already contributed to lowering several barriers to musical expression. The growing field of ADMIs ofers further promise, demonstrating that technology can play a transformative role in promoting inclusivity in musical practice. It is within this framework that the present research project is situated. The aim of UniMIDIHub is to broaden the possibilities for musical interaction without fundamentally altering the daily practices of its users. By enabling control through devices and tools that users are already familiar with, the system seeks to support creative expression while minimizing the cognitive and technical load often associated with new technologies. The observational studies carried out, first with individuals with diverse impairments, and later with students in the field of music technology, have provided valuable feedback that continues to shape the development of the system. The results confirm that, despite initial uncertainties, users can navigate and operate the software independently. Furthermore, live ensemble sessions involving both professional musicians and users with disabilities have ofered promising insights into the practical efectiveness and musical potential of the system in real-world settings. Looking ahead, this work contributes to the broader conversation on inclusion in digital music-making and ofers a concrete step toward the development of tools that accommodate diversity without compromise.
The authors wish to express their sincere gratitude to Audio Modeling for their collaboration in the co-development of UniMIDIHub and for their valuable technical support throughout the project. Special thanks are extended to Manuele Maestri, founder of Musica Senza Confini, for his insightful feedback and for enabling the organization of live performance sessions involving both professional musicians and users with disabilities. The authors also gratefully acknowledge the Laboratory of Music Informatics (LIM) at the University of Milan for their continued support and essential contributions to the research and development process.
Either: The author(s) have not employed any Generative AI tools.