What inspired us is that sometimes when giving a long presentation, you want to just jump straight to certain slides. Whoever said that slides had to be created in the order that they are to be used in? Only the restrictions of not having a solution like this one. Also, sometimes you want to give a presentation without a clicker, or simply don't have one. How annoying is it to go back to the computer to click each time? Why not have the computer switch slides for you?

What my colleagues learned is JavaScript, git, JSON, and python, especially how to use python to parse JSON and how to make a server with python. I learned how to use speech recognition api.

We built with github collaboration (until we got confused and just started using google drive instead). How the program works is a bit complicated, but basically the speech is converted to text, and that text is compared to some text that is already mapped to some slides. Ideally, that step is done with a user interface of some sort, but we didn't finish that part.

Challenges we faced are that microphones work differently for each computer, so we basically weren't sure if the api was working, and even when it did work, it didn't work nearly as well as expected. There was confusion over how the whole server architecture would work (php AND python? How?). Most significantly, there was the problem of how to choose between conflicting slides, as sometimes what the presenter says is actually associated with numerous slides, not just one, so how would the program choose which slide to show?

Built With

Share this project:

Updates