In order to enhance people’s ability to interact with their acoustic environments, hearing devices are common tools. However, it is difficult to evaluate the benefit of those tools or to measure acoustically challenging situations in natural environments. This paper proposes a way to measure the most important features of everyday acoustics environments by extracting a limited set of features while not compromising the privacy of partners and bystanders. The respective national laws on how to deal with audio privacy are very different among countries. The authors proposed using a smartphone as the source recorder but splitting the feature extraction into two phases: an initial feature processing in the smartphone and a later processing on a more powerful computer. For a given feature set, a statistical analysis shows comparable results from the extracted data when using either the original audio or the new privacy-aware extraction methods. A comparison shows that different scenarios result in separable features using the new extraction method.
(See document for exact affiliation information.)
https://aes2.org/publications/elibrary-page/?id=18298
(1134KB)
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member Join the AES. If you need to check your member status, login to the Member Portal.
