top of page

911 Pizza

We watched videos in which women called 911 to seek help by pretending to order pizza. In some cases, the dispatchers understood the code. In many cases, the dispatchers did not and hung up the phone. We created 911 Pizza to translate these cryptic Pizza orders and calls 911 describing the caller's situation.

Demo website

https://911-pizza.vercel.app/

Built With

​

  • speech-recognition

  • JavaScript

  • Node.js

  • Twilio

  • Vercel

Github

https://github.com/mariomanalu/911-pizza

Collaboration

Mario Manalu

Stephanie Cho

Technologies Used

  • We brainstormed the idea.

  • We did more research.

  • Mario started building the bare-bone of the web app.

  • Then, Stephanie worked on the front-end while Mario worked on the SpeechRecognition API.

  • Mario finished implementing SpeechRecognition and started implementing Twilio.

  • Stephanie made the landing and the "menu" pages.

  • Mario coded the pattern-matching algorithm to convert cryptic messages into nice, voice reports to 911.

  • We tested it out.

Hackaton Creatica 2022 Awarded Third Place Hack

What it does

  • 911 Pizza uses the SpeechRecognition API to receive the user's 911 Pizza order through voice recognition and convert it to text. Then, it pattern-matches the user's order and replaces some cryptic messages into easy-to-understand crime reports. After that, it calls 911 via the Twilio API and tell the police about the user's situation.

​

  • 911 Pizza has a "Menu" that contains a set of Pizza terms used to describe the user's dangerous situation. The "Menu" is updated and randomized to prevent abusers from knowing the meaning of 911 Pizza orders.

1.png
2.png

Challenges we ran into

  • We initially wanted to use Google Cloud's Speech-to-Text app. We spent 4 hours researching and trying to make it work but we could not. We could not understand how to connect Google's virtual terminal to our local terminal.

​

  • When we connected the node back-end to the front-end, there was a problem with the Twilio API. We were not able to send SMS. It turned out that the problem was we were using a VSCode Plugin to run our server, and it did not register the request.

​

  • Time constraint was also a challenge because we had a lot of homework and lived in different timezones.

Accomplishments that we're proud of

We initially wanted to use Google Cloud's Speech-to-Text app. We spent 4 hours researching and trying to make it work but we could not. We could not understand how to connect Google's virtual terminal to our local terminal.

When we connected the node back-end to the front-end, there was a problem with the Twilio API. We were not able to send SMS. It turned out that the problem was we were using a VSCode Plugin to run our server, and it did not register the request.

​Time constraint was also a challenge because we had a lot of homework and lived in different timezones.

What we learned

  1. We learned a lot about full-stack web dev with Node.js.

  2. We learned about http requests.

  3. We learned about how to prettify front-end designs.

  4. We learned about how to implement Twilio Calls in Node.js.

  5. We learned abot how to use SpeechRecognition.

What's next for 911 Pizza

  1. We will build a mobile app version of 911 Pizza.

  2. We will limit the number of orders to prevent users from spamming 911.

  3. Right now, we hard-coded the randomization of the "Menu". In the future, we want to automate this process.

bottom of page