A method to analyse and test the automatic selection of hearing aid programs
Hendrik Husstedt, Simone Wollermann, Jürgen Tchorz · Proceedings of the International Symposium on Auditory and Audiological Research · 2017
Digital hearing aids usually provide different hearing aid programs. This means different settings can be selected to adapt the signal processing to different hearing situations. Furthermore, advanced devices often include a classification algorithm that continuously analyses the acoustic environment and automatically selects a hearing aid program accordingly. However, there exists no method to analyse this adaptive feature. Therefore, we present a possibility to analyse and test which hearing aid program is active in a specific hearing situation. To proof the concept, hearing aids of two different manufacturers are analysed. These results give insights into the differences between classification strategies and classification quality among hearing aid manufacturers. Moreover, it shows that some signals, which humans can easily classify, are difficult to classify for hearing aids. Furthermore, the result of one device is compared with the classification entries of the data logging feature, which shows good agreement and verifies the new method. In addition, this comparison shows that the new method allows for a more comprehensive analysis so that using the data logging is no reasonable alternative.