Emotion-Aware Music Recommendation and Exploration
EmoMTB is an audiovisual interface to explore large music collections. It adopts the metaphor of a city, where similar songs are grouped into buildings. Nearby buildings form neighborhoods of similar genres. Users navigate through the city, exploring different musical styles either within their comfort zone or outside it. At the same time, an underlying AI monitors textual user-generated content to predict emotional states and adapts the audiovisual elements of the interface accordingly. Tailoring the results of a recommender engine to match the affective state of the user, EmoMTB provides a new way to discover music. In addition to this, EmoMTB encourages discussion on the capabilities of current machine learning algorithms to predict personal information such as emotion or personality traits, based on a user’s (or society’s) “digital footprint”.
Impressions of the Project “Music Tower Blocks” by Markus Schedl (AT), Florian Fritzl (AT), Franz Schubert (AT), Alessandro B. Melchiorre (IT), Oleg Lesota (RU), Emilia Parada-Cabaleiro (ES), Vasco B. R. Fragoso (PT), David Penz (AT), Christian Ganhör (AT) at the “Digital && Life” Exhibition. Credit: vog.photo
Music Tower Blocks (EmoMTB), Markus Schedl (AT) Florian Fritzl (AT) Franz Schubert (AT) Alessandro B. Melchiorre (IT) Oleg Lesota (RU) Emilia Parada-Cabaleiro (ES) Vasco B. R. Fragoso (PT) David Penz (AT), Credits: Santa Pile, Franz Schubert, Markus Schedl
Music Tower Blocks (EmoMTB), Markus Schedl (AT) Florian Fritzl (AT) Franz Schubert (AT) Alessandro B. Melchiorre (IT) Oleg Lesota (RU) Emilia Parada-Cabaleiro (ES) Vasco B. R. Fragoso (PT) David Penz (AT), Credits: Santa Pile, Franz Schubert, Markus Schedl
Music Tower Blocks (EmoMTB), Markus Schedl (AT) Florian Fritzl (AT) Franz Schubert (AT) Alessandro B. Melchiorre (IT) Oleg Lesota (RU) Emilia Parada-Cabaleiro (ES) Vasco B. R. Fragoso (PT) David Penz (AT), Credits: Santa Pile, Franz Schubert, Markus Schedl
Music Tower Blocks (EmoMTB), Markus Schedl (AT) Florian Fritzl (AT) Franz Schubert (AT) Alessandro B. Melchiorre (IT) Oleg Lesota (RU) Emilia Parada-Cabaleiro (ES) Vasco B. R. Fragoso (PT) David Penz (AT), Credits: Santa Pile, Franz Schubert, Markus Schedl
ZurückWeiter
M. Schedl: M. Schedl is a professor at JKU Linz and LIT, leading the Multimedia Mining and Search (MMS) and the Human-Centered AI (HCAI) groups.
E. Parada-Cabaleiro: E. Parada-Cabaleiro is a postdoc researcher at JKU.
A. B. Melchiorre & O. Lesota: A. B. Melchiorre and O. Lesota are PhD students at JKU.
D. Penz: D. Penz is a PhD student at JKU and TU Wien.
F. Fritzl: F. Fritzl is a Master’s student at FH Salzburg.
V. Fragoso & C. Ganhör: V. Fragoso and C. Ganhör are student research assistants at JKU.
F. Schubert: F. Schubert is a senior lecturer at the University of Applied Arts Vienna and the FH St. Pölten.
Credits
EmoMTB received financial support from the Linz Institute of Technology (LIT). We would also like to thank Michael Mayr and Peter Knees for their contributions to the first version of the prototype system, and to Antonia Ebner and Stefan Brandl for providing technical support.
