← ALL NEWS

DIREC · 11 Nov 2026 · DENMARK

From Cremona to Bang & Olufsen: Italian researcher Alessandro followed the sound to Denmark

Alessandro Cerioli, an Italian postdoc at DTU, is testing whether headphones work better with a team of small AI specialists than with one large model. He is part of REINS, a DIREC project carried out with Bang & Olufsen. He trained as a computer engineer at Politecnico di Milano, then came to Denmark for a master's in sound and music computing at Aalborg University. He later worked as an embedded developer at the hearing company GN and did an industrial PhD with Jabra and DTU.

The problem is that headphones must enhance speech as the wearer moves from a quiet office to a busy street, react instantly, and last all day on a tiny battery. A large, general AI model can handle any situation but is too big to fit in an earbud. It is also too slow if every sound has to go to a data center and back. Cerioli describes the large option as the "global solution": one model that is "less expert" but covers everything. The alternative is several compact tinyML models, each trained as an expert for one context. As he puts it, the more you improve quality, the more hardware resources you usually need, so the task is finding the best trade-off.

He sees the issue as wider than headphones. Most current AI is built big and uses a great deal of energy, and he considers this unsustainable financially and in terms of energy and water. AI could instead run in many more places, closer to where things happen, using far less energy. Finding where going big pays off and where going small is enough is the first step.

The work relies on edge AI, which runs on the device itself rather than in a data center. This saves energy, removes network delay and works without an internet connection. It also relies on adaptive AI, which switches to a different specialised model when language, noise or the task changes.

REINS runs from 2025 to 2027 and aims to make AI run efficiently on small, low-power devices. Its partners are DTU, the University of Southern Denmark, the Alexandra Institute, Bang & Olufsen and Leica Geosystems.

Read the original ↗