Inspiration

One of the presentations at the hackathon's opening ceremony where the text in the presentation slides were in sync with the presenter's speech. The combination was impactful.

What it does

As a user presents his/her speech, the web app converts the speech into texts and images in real time and fill the screen in a pseudo-presentation slide.

How we built it

We used Microsoft's Bing Speech API with a combination of JavaScript (Node.js, jQuery), HTML5 and CSS3 to build the web app from a Microsoft's sample client's library (https://github.com/Azure-Samples/SpeechToText-WebSockets-Javascript).

Challenges we ran into

None of us were very fluent with JavaScript. Had to do a crash-course..

Accomplishments that we're proud of

The web app works! It definitely surpassed our initial MVP expectations, but there are still bugs to be solved and more improvements to be made.

What's next for Prezo

1) Allow users to upload their images to complement what they want to say. 2) Allow users to select certain keywords they wish to emphasize/have special effects on. 3) Use NLP to process the speech and generate slides containing key messages and related graphics.

Just imagine a future where presenters do not need to prepare any presentation slide (well they have to memorise their script or be impromptu =P)!

Share this project:
×

Updates