TLDW

• 

  Home Page

Inspiration

A few different views: Paying attention in class and being able to learn can be hard. Having to jot down everything the professor is saying, while still taking time to comprehend it, can sometimes be unmanageable. I wonder if something can help...

The school hired a note taker for me because I either can't attend class or have a disability preventing me from taking my own notes. Unfortunately, the notes I get are either too long, or I feel like they are missing something. I wish I had a proper tool to help me be more independent and get amazing notes.

There is so much covered in my meetings, and they drag on forever, that I have trouble remembering the important topics that were covered. It would be great if there was a tool to help me keep track of what was going on.

What it does

TLDW: Too Long, Didn't Write is a fully featured web application that makes note taking and management a breeze. With an integrated Speech to Text function, let the software take the notes for you so you can spend your time paying attention and trying to learn.

For when you do need to take over, it contains an amazing text editor that allows for the proper formatting you need for even your most mathematical classes.

Finally, it can be difficult to figure out the key points in a lecture or meeting. Using advanced technologies with Natural Language Processing, check out the Summary tab to see the key points in your document.

With a simple to use, simple to learn UI, there is nothing stopping anybody from never having to worry about notes again.

How I built it

This application was built using Angular, and using Stdlib to enable a server-less architecture. A general Stdlib function was created to enable the general use of a text-summarization API. This function requires an api key (from smmry.com) and text to summarize, so it could be used by anybody, anywhere.

We set out to create an application that would be fully functional by the end of the Hackathon, and we definitely accomplished that!

Challenges I ran into

We had many more plans for the speech recognition and different features, but there were issues using various APIs. For example, Google's Speech To Text API requires single channel audio, and is only available for specific formats. Luckily, Mozillla has a defined standard for dealing with this when recording audio. Unfortunately, it hasn't been adopted by most browsers yet.

Accomplishments that I'm proud of

We actually got it working with realtime saving and being able to have multiple users logged in at the same time, all working on different things.

What I learned

It is very important to thoroughly investigate an API when you wish to use it. A good way to find out how to use it is to search on Github for other users that have implemented it in their projects, especially when the API lacks good documentation.

What's next for TLDW

We would like to add sharing and allow multiple users to work on the same document. We would also like to experiment with different Speech to Text APIs to test accuracy, and maybe even combine the results to have it as accurate as possible. As well, it would be great to have the recorded text be linkable to the audio, so if you question what was turned to text, you can investigate the audio file yourself.

Built With

• angular-5
• stdlib