Master Theses - Details

Supression of Wind Noises in Hearables and Hearing Aids

Supervisor:  Christoph WeyerJohannes Fabry

Area: Audio Signal Processing, Active Noise Cancellation, Acoustic

Tools: Matlab, Acoustic Measurements, Sigma Studio

Category: Master's Thesis

Status: Open

Hearables and hearing aids integrate more and more signal processing to achieve active noise cancellation and features like smart hear-through. To measure the necessary signals, the devices are equipped with outer and inner microphones. A big problem with these devices is the occurence of wind noises.

Wind creates disturbances at the microphones, which result in very loud noises.In this thesis, different approaches to the problem shall be examined.

In a first step the problem sall be quantified by measurements on commercial headphones. The resulting measurements could in a next step be compared with methods for the simulation of wind noises, as known from Nelke or Habets. Afterwards, methods for the supression of wind noises shall be analyzed by simulation. Classic methods include the detection of wind noises and the application of filter to surpress the affected regions of the spectrum. Also, adaptive filters or architectures employing multiple microphones would be possible. In a last step, the methods shall be implemented in a real-time system and evaluated by acoustic measurements.

The thesis is done in cooperation with a startup from the institute.