{"id":5792,"date":"2026-05-05T09:00:00","date_gmt":"2026-05-05T07:00:00","guid":{"rendered":"https:\/\/hacnumedia.org\/?p=5792"},"modified":"2026-05-05T09:09:02","modified_gmt":"2026-05-05T07:09:02","slug":"como-the-tiny-model-that-generates-sound-through-movement","status":"publish","type":"post","link":"https:\/\/hacnumedia.org\/en\/como-the-tiny-model-that-generates-sound-through-movement\/","title":{"rendered":"CoMo, the tiny model that generates sound through movement"},"content":{"rendered":"\n

In this new series, HACNUM\u00e9dia invites readers to explore the tools shaping contemporary digital creation. Not only the most visible ones, but also those circulating at the margins of dominant trends. Here, the focus is on CoMo: a web-based environment developed at IRCAM that uses small machine learning models to link gestures to sounds. Technically, it is AI. In practice, it has very little to do with what that term commonly evokes today. <\/strong><\/p>\n\n

Following our article<\/a> on tiny models and ecological alternatives to generative AI, CoMo<\/a> illustrates another possible approach: frugal, embodied, collective. Developed for nearly ten years by IRCAM\u2019s<\/a> ISMM<\/a> team, the tool remains relatively discreet, known mainly within circles of musical research and experimental sound creation. <\/p>\n\n

What is it for? Playing sound through movement <\/strong><\/h2>\n\n

\u201cCo\u201d stands for collective. \u201cMo\u201d stands for movement. Two words that capture what CoMo aims to do: enable groups of people to play with sound together by moving. \u201cThe main idea was to do things collectively. Not just an interface for a single user, but a group of people interacting together,\u201d explains Fr\u00e9d\u00e9ric Bevilacqua,<\/a> Research Director at IRCAM and head of the ISMM team (Interaction Sound Music Movement).This focus on gesture as a sonic interface is not new: the ISMM team has been working on gesture-based sound control since the mid-2000s. CoMo<\/a>, whose earliest versions date back to 2017, builds on this lineage. The tool is accessible directly through a web browser, requires no installation, and is open source. For motion sensing, no specialized hardware is needed: a smartphone is enough. Not for its screen\u2014deliberately diverted from its usual function\u2014but for its built-in accelerometers and gyroscopes. The different versions of the application, mainly developed by Benjamin Matuszewski <\/a>(CoMo Elements, CoMo Vox<\/a>, CoMo.Education<\/a>, and CoMo Rehabilitation<\/a>), share the same core engine: interfaces adapted to different contexts of use. <\/p>\n\n

<\/p>\n\n

\"\"<\/figure>\n\n

<\/p>\n\n

What makes it distinctive: interacting with the body, not with prompts<\/strong><\/h2>\n\n

CoMo uses machine learning techniques. But calling it \u201cAI\u201d in 2026 is almost misleading. \u201cNow, compared to the popular imagination, we can\u2019t really call this AI anymore,\u201d acknowledges Fr\u00e9d\u00e9ric Bevilacqua. large language models<\/a>CoMo belongs to what researchers call interactive machine learning: more traditional, lightweight models that can be trained with very little data\u2014sometimes a single gesture\u2014in just a few seconds, directly on a smartphone processor.\u201cWithin the realm of small data, these are really the smallest. Tiny, tiny models, <\/em>\u201d Fr\u00e9d\u00e9ric summarizes. This is intentional: the term \u201cAI\u201d was deliberately avoided so as not to distance audiences drawn to gesture, the body, and music. Another key distinction between CoMo and current AI tools lies in the role of training within the process. There is no separate phase: you record a gesture, associate it with a sound, test it, adjust it. All of this unfolds within the same creative gesture.\u201cTraining is fully integrated into the design. At any moment, you can record or modify. It\u2019s extremely flexible,\u201d <\/em>explains Fr\u00e9d\u00e9ric.<\/p>\n\n

<\/p>\n\n

\"\"<\/figure>\n\n

<\/p>\n\n

How does it work? <\/strong><\/h2>\n\n

In practice, using CoMo comes down to four steps:record a gestureassociate it with a soundtestrefineRepeat as many times as needed. The workflow is intentionally minimal\u2014and that is precisely where the tool\u2019s strength lies.To get started, there are two scenarios. The simplest is to connect directly to the version hosted by IRCAM, accessible through any browser. No installation, no account required. The available sounds are predefined and the session is temporary, but this is more than sufficient for a first workshop or quick introduction.The second scenario, for those wishing to work with their own sounds, involves installing CoMo on a local machine. This requires some technical familiarity\u2014modern browsers now impose secure HTTPS protocols to access sensors, among other constraints\u2014although a new application currently under development should soon simplify this step. Once installed, a simple Wi-Fi router is enough to create a local network to which smartphones can connect. For practitioners already equipped with creative tools, bridges with Max<\/a> \/ Max For Live<\/a>, TouchDesigner<\/a>, or other creative development environments are also being prepared. <\/p>\n\n

\n