The need to combat the isolation that can come from digital technology, combined with the need to help minority groups improve their english, resulted in PRO-nounce.
What it does
The web app allows users to record audio. Each file is passed to the server where Azure Speech Recognition is used to convert the speech to text and analyze it against the word of the day. If it matches, the user gains a point to their score. The leaderboard keeps track of the scores of all the users. At the end of the day the top 50 users get regular stickers to recognize their achievement, the user in 3rd place gets a bronze sticker, 2nd place gets a silver sticker, and 1st gets a gold sticker.
How I built it
Using Keystone - a model driven content management system, we created a backend to manage the site as an administrator. After that I worked on styling the frontend using HTML and CSS. I also integrated with Azure Speech Recognition to validate the pronunciation of the words and I store all recordings on Azure's Storage Container. Finally, I hosted the web application on an Azure Virtual Machine.
Challenges I ran into
I didn't have enough time to implement all the features I wanted to implement. For example, Azure's Speaker Recognition/Identification of the recordings to ensure each recording from a user is coming from a unique person so that users don't record one person multiple times and get points for that. One of the main aims of the app is to combat social isolation, this is difficult to achieve if the user isn't forced to go out and find multiple DIFFERENT people to record. In addition, one of the main features that must be implemented is the ability for the user to edit their information. I also ran out of time but will hopefully implement both of these features and more.
Accomplishments that I'm proud of
Being able to come up with a unique game idea and execute a very large chunk of it in a short period of time. I've also never been responsible for a website's backend, so I'm proud I was able to get it up and running.
What I learned
I explored the usage of Azure Cloud Services and learned how to work with audio files. I used FFMPEG to convert the audio files recorded by the browser to the format Azure Speech-To-Text recognizes (ogg).