Detection of Social Engagement Interaction using Smartphones

| 0

Type: Bachelor Master UROP
Status: Available

Recent improves in mobile technologies (better and richer sensors, cheaper and faster storage) allow us to record different aspects of our daily activities, e.g. work meetings, discussions during coffee breaks, family moments, etc. Within the EU research project RECALL we are investigating how such data captured from mobile and wearable devices can serve as “memory cues” to help individuals remember certain past events – e.g., a set of images can help one remember an evening with friends or pictures showing white board content can help one remember what was discussed in the last work meeting.

We are interested to understand how such data can be safely shared with other co-located people – e.g. with all the colleagues that participated in the meeting – in order to construct a better representation of what really happened in an event. Hence, our bigger goal is to build a solution for peer2peer exchange of captured memory cues or moments.
The goal of this bachelor project is to develop an Android mobile app that uses audio data (and/or other mobile sensors) to detect when two people are socially interacting (e.g. speaking). This will be used as a trigger to decide when to start our peer2peer moment sharing. If time permits the student can also experiment and try to construct a solution for detecting the presence of nearby peers using audio signals.
The project requires solid Android programming skills, preferably some knowledge on how to work with audio data and willingness to quickly learn new things.

For more information contact: