Lucerne University of Applied Sciences and Arts (HSLU) and the hearing aid manufacturer Sonova are working together to gain a better understanding of sound propagation in rooms. According to a statement, data is being collected in various rooms and situations as part of a student project. Sonova is contributing its knowledge of hearing acoustics, audio technology, signal processing and electroacoustics. HSLU can offer expertise in robotics, software development, room acoustics, and audio signal analysis and processing. The project is being funded by the Innosuisse innovation agency.
"If these detailed measurements are carried out manually by humans, they are extremely time-consuming," explained HSLU acoustics expert Prof. Dr. Armin Taghipour. This is why he and his team have developed mobile robots that can collect the data autonomously. Unlike earlier models, the robots can independently measure both the volume of background noise and the room acoustics.
The data generated in this way will serve as the basis for algorithms that can use machine learning to simulate different room situations and furnishings. “The most important prerequisite for machine learning to work is large amounts of data,” added Hannes Wüthrich, Project Manager at Sonova.
Unlike written texts or images, where existing data sets can usually be used, acoustic data must first be generated. Ultimately, this project aims to improve the quality of hearing aids, in situations such as reverberation in large rooms or loud background noise. This will make communication easier for people with hearing loss.
