Alberto Ancilotto from FBK participated at EUSIPCO 2022 (29 August – 2 September) in Belgrade, Serbia. EUSIPCO is one of the most important European conferences on signal processing.

MARVEL organized a special session on “Edge-Fog-Cloud Machine Learning for Smart Cities Applications”, where Alberto presented “Optimizing PhiNet architectures for the detection of urban sounds on low-end devices”. This work describes some of the efforts we are making in MARVEL to bring intelligence to the edge. The targeted task is Sound Event Detection (SED), which can be helpful in smart cities context to capture relevant events and generate alerts or turn on other sensors (e.g. cameras) to interpret better what is happening in a public place.

During the conference, Alberto explained how we applied a hardware-aware scaling approach (using the PhiNets) to perform SED on IoT end devices. The poster was at the same time a demonstrator. It had a special hole where a prototype of the SED running on a microcontroller was inserted and showed the recognition of sounds in real time!

Alberto commented: “I am honored to have been able to present my work at the 30th edition of EUSIPCO, in Belgrade. It really was an amazing experience, and being able to discuss my work with people working in the same field has sparked many interesting ideas. Special thanks to the organizers and to everyone who showed up!

The article will soon be online on IEEExplore. In the meantime, you can find a pre-print here

For more direct updates regarding our project as well as the topics we are studying and more, follow us on LinkedIn and on Twitter.

For regular updates including a collection of news and relevant information sign up for our Newsletter below.



Key Facts

  • Project Coordinator: Dr. Sotiris Ioannidis
  • Institution: Foundation for Research and Technology Hellas (FORTH)
  • E-mail: 
  • Start: 01.01.2021
  • Duration: 36 months
  • Participating Organisations: 17
  • Number of countries: 12

Get Connected



This project has received funding from the European Union’s Horizon 2020 Research and Innovation program under grant agreement No 957337. The website reflects only the view of the author(s) and the Commission is not responsible for any use that may be made of the information it contains.