Inspiration

Facebook is an unordered collection of vast amounts of data. Sometimes, due to its vastness, it is hard to keep track of useful information unless you are always keeping track of it all the time. We wanted to provide an easier way to keep track of what's happening in Facebook. By interacting with the assistant, you will know what your friends are up to.

What it does

Provides a voice interface allowing you to efficiently collect intelligent information from your unordered facebook data, such as whether any of your friends have been to a particular location, or whether any specific friend has seen a particular movie, etc. It only provides information from your facebook friends who have joined the app.

How we built it

We used Visual Studio to deploy on the app server. We created a backend database on Azure, and connected it to the app server. We used guidelines from Azure to setup the Android mobile application, that connects seamlessly with the app server, and directly connects with Facebook. We set up Facebook authentication on Azure. We used luis API as well as Bing Speech api for services.

Challenges we ran into

Using luis (Microsoft Cognitive Services). Since we have not used this before, there was a learning curve involved in figuring it out. Mobile App Development is also something new to us, so that was an additional challenge. Another challenge was connecting Android speech to text and luis in an efficient way.

Accomplishments that we're proud of

Managed to integrate applications from scratch such as Android and MVC5 with Azure. We integrated these applications from scratch for flexibility.

What we learned

How to use luis and Android Speech-to-Text. How to deploy using Microsoft Azure, and create scalable websites.

What's next for FacebookAssistant

We want to make the voice interface more interactive, and provide a way for it to learn more speech. We also want to provide more useful information to the user.

Built With

Share this project:

Updates