CONNEXOUNDS
Toon van Waterschoot
Toon van Waterschoot

Cooperative Signal Processing Solutions for IoT-based Multi-User Speech Communication Systems

In this project, we aim to develop multi-user speech communication systems exploiting the connectivity between multiple speech and audio devices in the Internet-of-Things (IoT). This will be done through the development and implementation of cooperative, and possibly distributed, signal processing algorithms, relying on recent research in the field of wireless sensor networks. In particular, we will consider multi-user Voice-over-IP conferencing, which is currently plagued with poor sound quality (due to noise, acoustic echo and feedback effects) and hence offers a poor user quality-of-experience. We will develop, implement and evaluate cooperative signal processing solutions for an ad-hoc conferencing system, i.e. a conferencing system involving the use of multiple wirelessly connected personal devices.

Period: Oct. 2019 – Jun. 2022

Funding: KU Leuven Industrial Research Fund (Category 3 Project), Project No. C3/19/013

Budget: 385 kEUR

Partners: KU Leuven

People:

  • Giuliano Bernardi (Postdoc)
  • Amin Hassani (Postdoc)
  • Marc Moonen (Supervisor)
  • Toon van Waterschoot (Co-Supervisor)

Website: connexounds.com