We sought to build an app to bridge the increasing technological divide between smartphones and phones that do not have internet capabilities. As a result, we came up with SMSiri -- a tool that enables users without internet plans on their mobile devices to access the valuable data of the web via text message / SMS.
What it does
SMSiri utilizes natural language processing to ascertain the intent of a SMS message, and then utilizes one of various APIs to answer the message and compose a reply.
How I built it
SMSiri is built on a flask backend, using a myriad of APIs. We used the wolfram|alpha API for general information retrieval, the Bing Maps API for step by step navigation between an origin and destination, the Microsoft Translator API to translate phrases between two languages, the Open Weather Map API to get the present weather at any location in the world, the Twitter Python library to retrieve the latest tweet of a certain user, the Yahoo Finance API to look up the prices of stocks, the Expedia Activities API to discover interesting activities near a location, and lastly the Microsoft Azure Bing News API, which allowed users of SMSiri to view headlines pertaining to any topic or location around the world through text messaging.
Challenges I ran into
After deploying to heroku, we realized that certain functionalities had been destroyed. As a result, we tested the code on local only to find it unfaulty. After several hours of troubleshooting, we were able to (with the help of a Twilio developer evangelist) ascertain that it was due to heroku's attempting to print out unicode characters.
Try SMSiri out at (714) 500-7689! A full list of commands / what it can do can be seen in the devpost app screenshots.