-
Class Selection Main Screen
-
Editing of Speech-to-Text Transcript
Inspiration
Keeping up with lectures can be difficult while having to constantly write down notes. This is especially true when considering international students (245,895 in Canada, 2017) and language barriers, or those with hearing issues. We aimed to reduce barriers and inefficiencies in learning through an accessible collaborative platform. With speech-to-text functions, we wanted to build an interactive and inclusive note-taking platform to make learning easier for everyone.
What it does
NoteScribe is a collaborative note taking web application that automatically transcribes recordings into edible text blocks. Notes can be organized by course lectures or other groupings. The transcripts can be modified, annotated, and shared for collaborative note-taking.
How we built it
Flask, Socket.IO, Firebase, Microsoft Azure AI API, Bootstrap & Javascript.
Challenges we ran into
Incompatibility between the Azure API and node.js made us change our plans to Python and Flask, which were less familiar and still had some implementation challenges with setting up our server.
Accomplishments that we're proud of
Some accomplishments we're proud of are the fact that we learned to use the Azure API as it was very difficult to learn how to use. We're also proud that we managed to accomplish as much as we did in 36 hours!
What we learned
We learned about object-orientated programming, how to use the Azure API along with implementing the speech to text functions through a server and web application.
What's next for NoteScribe
Improving input accuracy and filtering out excess noise through potential training/machine learning with the speech to text API would allow better functionality in noisier settings. Additional editing/annotating options and live recording are all areas for improvement. ReactJS could also be used instead for an easier to manage front-end.
Log in or sign up for Devpost to join the conversation.