Host: Francesco Sabaini, MoveByBIke Italy, Italy

Visitor: Joshua Ng, Sonic Assembly, Sweden

Velosonics is a conceptual project exploring the feasibility and applications of sonic data collected through contact microphone recordings of urban delivery bikes. Its primary goal was to test a methodology for capturing and analysing data using both conventional and experimental sound design techniques. A secondary aim was to evaluate tools for automating this process and to identify practical use cases for the findings.

The project represents a cross-sector collaboration between Swedish creative sound studio Sonic Assembly and Italian last-mile logistics company MoveByBike Italy, driven by a shared belief in data-led design and curiosity about unconventional approaches to data collection.

Project Results

The initiative took place in Verona, Italy, in May 2025. Velosonics developed a proof-of-concept machine learning model for monitoring MoveByBike’s Laplander cargo bikes, accompanied by a React-based code package integrating Google Teachable Machine/TensorFlow, enabling others to build similar “smart ear” systems.

The collaboration combined expertise in sound design, engineering, urban logistics, software development, and media production. Key outcomes included over five hours of delivery bike acoustic data and insights into how sound analysis can support both operational planning and product design. Engineers expressed interest in using spectrogram analysis to address performance evaluation gaps in current simulation software for cargo bikes.

Looking forward, Velosonics aims to evolve this prototype into a standalone, customizable device and software platform, empowering users to train their own models and acoustically monitor a variety of objects.

Further details, including methodology documentation and a 6.5-minute explainer video, are available on the project’s dedicated webpage.