Listening to energetic music to push yourself while doing your workout? Definitely. Being more willing to dance on a friday night than on a monday morning? True. Feeling moody on a rainy day? Oh, yes!

It fascinates me how much power contextual information such as location, time, weather, activity and emotions can have on the choice what kind of music to listen to. Thanks to the various mobile sensors in wearable devices, we can track where we go, how many steps we take, how well we sleep and so much more. I strongly believe that these sensor technologies could also be beneficial for music streaming services in order to play the right music in the right moment automatically.

For this reason, I have immersed myself into the field of music information retrieval for my bachelor thesis and explored several sensor technologies that can help to capture this information.The presented system takes sensors such as accelerometer, heart rate sensor, barometer, microphone sensor and camera sensor into account. Moreover, technologies like object and facial emotion recognition, musical feature processing, the extraction of relevant keywords from the calendar and the data collection from sleep applications are considered as well. 

In the next step, I presented the concept for a context-aware music recommender system and in doing so, my main focus was the user interface and the way music listeners interact with it. 

The final result is a mobile user interface that adapts itself visually to the user’s context. It is expected to intensify the listener's mood by visualizing captured contextual information such as the user’s location, time of day and weather on the user interface. In addition, extracted musical features  of the currently playing song, including danceability, valence and  energy, influence the appearance and animation of the play button.

Bachelor Thesis – For more details visit:
https://www.researchgate.net/publication/333903761_Konzeption_und_Visualisierung_eines_User_Interfaces_eines_kontextsensitiven_Musikempfehlungssystems_zur_Verbesserung_der_User_Experience_beim_Musikhoren