Inspiration

The inspiration for InstaPresent came from our frustration with constantly having to create presentations for class and being inspired by the 'in-game advertising' episode on Silicon Valley.

What it does

InstaPresent is a tool that uses your computer's microphone to generate a presentation in real-time. It can retrieve images and graphs and summarize your words into bullet points.

How we built it

We used Google's Text To Speech API to process audio from the laptop's microphone. The Text To Speech is captured when the user speaks and when they stop speaking, the aggregated text is sent to the server via WebSockets to be processed.

Challenges We ran into

Summarizing text into bullet points was a particularly difficult challenge as there are not many resources available for this task. We ended up developing our own pipeline for bullet-point generation based on part-of-speech and dependency analysis. We also had plans to create an Android app for InstaPresent, but were unable to do so due to limited team members and time constraints. Despite these challenges, we enjoyed the opportunity to work on this project.

Accomplishments that we're proud of

We are proud of creating a web application that utilizes a variety of machine learning and non-machine learning techniques. We also enjoyed the challenge of working on an unsolved machine learning problem (sentence simplification) and being able to perform real-time text analysis to determine new elements.

What's next for InstaPresent

In the future, we hope to improve InstaPresent by predicting what the user intends to say next and improving the text summarization with word reordering.

Share this project:

Updates