Crowdsourcing speech data through mobile applications is relatively new. In the present contribution we add to the existing body of research an innovative Android and iOS application called `Voice Äpp'. The free app is pioneering in the sense that it leverages its function as a medium for science communication — thus attracting an extensive user base — to crowdsource audio and dialect data. The app was launched in early 2015 and has already been downloaded 19k times. Nearly half a million audio tokens have been crowdsourced. In this system levels contribution we describe the basic functionalities of the app — voice and dialect analysis —, we present the scientific potential of the corpus created, and discuss methodological issues related to crowdsourcing audio data through mobile applications.