Music Context Player

This mobile application developed for Nokia N9 is a system for collecting context and usage data from mobile devices, but targeted at recommending music via unsupervised learning of user profiles and relevant situations. The developed data flow system supports both short enough response times and longer asynchronous reasoning on the collected data; furthermore, the mobile phone acts not only as sensor, but the mobile app is directly tied to the effectiveness of the music service user experience (UX).

My role on the project: Interaction design
Collaborators: Mauricio Desari (Interaction design), Glaubert Souza (Motion design), Anderson Schimuneck (Visual design), Francimar Maciel (Usability research), Tamara Baia (Usability research)
Platform: MeeGo OS, v1.2 Harmattan
Mobile: Nokia N9
Microsoft research publication: http://research.microsoft.com/apps/pubs/default.aspx?id=204231

Music listening is a very personal and situational behaviour, which suggests that contextual information could be used to greatly enhance music recommendation experience. However, making such use of mobile context, while learning user profiles, is a challenging problem. 
 
This mobile application developed for Nokia N9 is a system for collecting context and usage data from mobile devices, but targeted at recommending music via unsupervised learning of user profiles and relevant situations. The developed data flow system supports both short enough response times and longer asynchronous reasoning on the collected data; furthermore, the mobile phone acts not only as sensor, but the mobile app is directly tied to the effectiveness of the music service user experience (UX).
 
I had the opportunity to work on this project as an Interaction designer with Mauricio Desari. Together, we developed an fun UI for music lovers, giving to then a intuitive and interesting application.
1.0) Persona: Tommi Martin
As an contextual application we had the necessity to create a persona to illustrate the recommendations in the screenflow. The persona was a heavy metal listener, which likes to discover new bands and analyze them. All app usage scenarios are designed around the concept of having the app just work and the system recommend songs related to the current user context, without demanding much effort from the user. When launching the app for the first time, the user is prompted to loads his/her collection of music files. Based on his actions, the application populates the background with cd covers and also filters the recommendations in the center. 
2.0) Tapping a recommendation
If the user taps a recommendation (song, new, store or event), he or she goes to a different context. If is a song, the user opens and the song is played right away. If is a new, store or event, a panel appears, showing different contexts that can be explored.
3.0) Contextual carousel
By tapping on any contextual recommendations, a carousel appears showing more relevant contexts to the user. He or she can explore them by flicking the carousel to the right or left. At the bottom of the screen, the user has free access to the song that has been playing, he can also change the song or stop.
4.0) Music player
The music player can also help with the recommendation. If the user press the heart, he or she is confirming that this song is favorite. The user can also dislike the song, by tapping the broken heart. 
Usability: Evaluation of the quality of a recommender system can be seen from three different points of view: functional testing, quality of recommender algorithm output, and usability evaluation. If you want to know more about the usability tests for this project, please, download the research on Microsoft Reseach website: http://research.microsoft.com/apps/pubs/default.aspx?id=204231
Back to Top