Composer - Performer - Researcher
Publications
Gesture Mapping for Embodied Rhythmic Expression: A Case Study on Expressive Affordances
Evan O'Donnell and Atau Tanaka
MOCO 2026
We provide a case study applying movement and muscle sensors to work with rhythmic phrase and pattern in digital music performance systems. While there is much prior research on gesture-sound mapping, comparatively little focuses on rhythm. However, strong correlations between rhythm and movement suggest productive untapped creative affordances. In this work, we detail the use of inertial measurement units (IMU) and electromyogram (EMG) signals to develop a series of interaction designs exploring the affordances of movement-rhythm mapping. We describe the use of an iterative, movement-led design process building on prior improvisation analyses and practice research methodologies to identify an intuitive gesture-rhythm performance language. We implement these insights through direct data mappings, machine learning via classification and regression, and timbre transfer using variational autoencoders. Our results demonstrate the broad expressive possibilities of this approach for creative practice, while outlining numerous productive paths for further inquiry.
Real-Time Gesture Classification via Multi-Modal Sensor Data for Intuitive Performance Mapping
Evan O'Donnell and Atau Tanaka
CMMR 2025
We present an approach to real-time continuous gesture recognition via accessible classification and regression tools and its application in an embodied musical performance workflow. By combining electromyogram (EMG) muscle signals with gyroscope and accelerometer data from a wireless inertial measurement unit (IMU) system, we attained well-rounded descriptors of ongoing gestural arm movements. By training on this data via multilayer perceptron classification and outputting confidence ratings in place of predicted classes, we successfully detected five pre-chosen gesture types in real time and interpolated smoothly between their associated audio clips. Integrating this model into an interactive performance system let us harness these confidence ratings as overlapping influences on a musical arrangement, expanding on embodied musical associations via intuitive aesthetic mappings. Our results demonstrate the feasibility of continuous gesture recognition with the current generation of accessible machine learning tools, extending prior research into new use cases while enabling exploration and application of embodied expressive associations.
A Practice-Based Methodology for Capturing Embodied Gesture-Rhythm Relations in Small Datasets
Evan O'Donnell and Patrick Hartono
AIMC 2025
This paper introduces a practice-centered methodology for building gesture and rhythmic phrase datasets to support music composition and performance. Rooted in individual investigations into gestural sensor use for sculpting rhythmic expression in electronic music, the approach integrates an improvisation analysis method developed by Rodrigo Constanzo with sound-tracing workflows. The resulting three-part dataset—comprising audio recordings, inertial measurement unit (IMU) and electromyogram (EMG) sensor data, and rich qualitative commentary—provides a flexible framework for merging embodied musical intuition with machine learning techniques. Positioned within the context of small-data AI, this work offers a nuanced, artist-centered contribution toward integrating qualitative and quantitative insights into ML-assisted systems for creative and expressive applications.
Presentations
DMRN+19: Digital Music Research Network One-day Workshop, London, UK, 2024: Oral presentation: "An improvisation analysis method with machine learning for embodied rhythm research."