Public speaking is one of those anxieties that everyone has. It is difficult to overcome, mainly because people lack the opportunity to get in front of a group of people to practice speaking. This lack of practice exacerbates the fear, creating a positive feedback loop, making public speaking the worst phobia in North America. We built SpeakEz with the hopes of helping people make public speaking a normal thing.

What it does

SpeakEz presents the user with the proper environment to practice public speaking. By utilizing a virtual reciprocating crowd, the user can prepare for any public speaking scenario. Users have the option of listening to their own voices (something many dread), as well as seeing their own hand movements. This allows the user to become comfortable with public speaking in private until he or she becomes a public-speaking pro!

How we built it

Our team utilizes Unity and the Oculus Rift API to provide the immersive environment for the user. To further provide a realistic feel, we attached the head-mounted display with a Leap Motion, allowing users to see their hands and the different gestures that people unconsciously make. Lastly, we used a 3D printer to create a compact kit that held the setup together.

Challenges we ran into

Our main challenge was figuring out different methods of measuring anxiety using the hardware and software we were provided with. Initially, we tried using a speech-to-text API to feed data into a tone analyzer, but after being unsuccessful in working with the APIs, we had to search for other queues that could be used to measure anxiety. Coordination was also a challenge, as using git would be near impossible with the sheer size of the files.

Accomplishments that we're proud of

This was the first time the team had worked with the Oculus Rift (or any VR dev kit for that matter). We also built an accurate model of the SDHacks event in the short period of time we were provided with. Lastly, we overcame software and hardware difficulties and were able to create our own algorithms for measuring how nervous users were.

What we learned

We learned how to cope with the hardware issues that made development hard.

What's next for SpeakEz

The team plans to work on making the models more realistic, while nailing down the version control for when we all work on the project.

Share this project: