Auto911

I've always wondered why there isn't a way to text the police. If an intruder broke into my house, It would be difficult for me to contact authorities in fear of being heard. It would be much easier to have a resource on hand which can alert the police while I stay silent. Auto911 allows those in danger to contact the police in a similar fashion to a basic voice call while staying completely silent. With the push of a simple button, the nearest police station will be called and alerted through a synthetic human voice the person's location as well as their emergency. This ensures maximum protection and safety. My team and I combined a variety of different resources to create this web application. To deploy our project, we used Google's App Engine. We also used Flask and HTML in order to build and enhance our website. The Google Maps API tracks the user's location and finds the nearest police station as well as its phone number for increased efficiency. These pieces of information, as well as an optional typed message by the user, are converted into an mp3 file with the Google Text to Speech API, with the words read out loud. Twilio is then used for the purpose of calling the number of the nearest police station and playing the mp3 file from the Google API on the phone call, simulating a real human voice. The way this works is the mp3 file is sent to Google Cloud Storage as a blop in a bucket and is received as a URL which Twilio calls on along with the TwiML bin to play the file. The police will then be alerted and ready to take action. My team and I ran into a wide range of challenges throughout every step of our project, with each of them taking hours of our time. We first ran into problems with the installation of python in the wrong folders, but once it was fixed, it had to be completely changed as some applications did not allow for the newer types of python but instead only the older versions. Both the Google Cloud APIs we worked with took a lot of breaking down and figuring out as we had never worked with them before and were confused as to some of the specific aspects of their implementation. It took a while for us to achieve a precise location of the police stations around us as well. The hardest and most time-consuming challenge was the implementation of the Twilio API as well as merging it with the Google Text to Speech API. The Twilio API required us to convert our mp3 file into a URL- it was difficult to find a platform that could not only do this but do it programmatically, and we ended up using Google Cloud Platform. Then our greatest problem arose, which was the issue of the mp3 files that Twilio would play. Instead of playing exactly what we wanted to, it somehow accessed old mp3 files and almost never played the current one but instead jumped around previous ones. We were very confused and tried 4 different ways of coding the program as well as 3 different solutions to the problem over 5 full hours, including deleting the storage buckets and deleting the cache history. My group and I are proud of the amount of work we put into our project despite the number of problems we ran into. We definitely were not expecting most of them and feel accomplished that we were able to persevere. We also feel glad that we had the opportunity to create a project that can help a very large amount of people and feel honored to do so. Every single application my group used was new to me, and I learned about the specifics about each of them. The APIs, which were confusing at first, were extremely interesting and I became more and more comfortable with them as the weekend went on. My group agrees as we were introduced to a lot of APIs that looked way more simple in the demonstrations than when we actually used them. We learned to have a newfound appreciation for all of these projects and how to use them to create amazing projects in the future. In the future, Auto911 can become a mobile app instead of a web application. It could also connect to a database and store more information.