Music Affect Data Collection App
Design an app to allow users to self-report affective (emotional) response to music. The app will use the user’s own Spotify account to play songs and the user will rate their response dynamically using a joystick-like widget.
Objectives
- Design a cross-platform mobile app (in Flutter) that collects continuous annotations of users affective responses to music.
- Integrate with Spotify API to stream music on the app without storing audio recordings.
- Maintain a cloud-based dataset of collected data.
- Validate product with a small and informal field study.
Motivations
The automatic estimation of human affective responses to music is a research task of great interest to the field of music information retrieval. However, copyright concerns and the cost of carrying out formal psychological lab studies, existing datasets are few in number and small in size. We seek to build a crowd-sourcing solution to enable to collection of large datasets of affective responses to music. By leveraging users' own Spotify accounts, we can bypass concerns about audio copyright and collect only data pertaining to the user response.
Qualifications
Minimum Qualifications:
- music lover
- CS 492 (mobile) or CS 493 (cloud)
Details
Project Partner:
Patrick Donnelly
NDA/IPA:No Agreement Required
Number Groups:1
Project Status:Accepting Applicants
Website:https://soundbendor.org/
Keywords:
Musicuser interfaceMobile AppFlutteremotionpsychologySpotify