Virtual Assistant UROP- FALL 2016

 

Mobile phone assistants, like Siri, are increasingly becoming our primary way to interact with our devices. Voice assistants are no longer limited just to mobile devices, as Siri is now on Mac with macOS Sierra and Amazon's Alexa built into the Echo. While they are coming close to human-like interactions, they know little about us or our daily activities. Imagine Siri waking you up every day and automatically turning your alarm off if you have already woken up, or continuing to wake you up every 10 minutes if it realizes that you are still in your bed. Think about how cool it would be for Siri to ask you to take a walk if you have been sitting too long?

 

This is a vision that can become reality with recent research from our group that has shown that it is possible to track people without any wireless device on them (http://witrack.csail.mit.edu/and http://witrack.csail.mit.edu/vitalradio/). This project will involve taking this research further by inferring activities from trajectory data and integrating this information with phone assistants like Siri. This is a full time fall UROP opportunity with potential to continue through IAP and the Spring semester.

Prerequisite skills: General programming skills (C/Java), iOS App Development experience (Swift / Objective C)

Skills that are helpful to know: Interacting with REST APIs, some background in Machine Learning

Commitment: 10-15 hrs/week, for credit or pay (Direct or UROP funding)

Contact: Please send your resume, unofficial transcript and a brief paragraph describing why you would be interested in working on this project to Deepak Vasisht (deepakv@mit.edu) and Anubhav Jain (ajain94@mit.edu).