In the past, a lot of our team members have struggled with public speaking or even speaking in general. The idea for this project actually came from one of our members who, despite doing few class presentations in the past, still feels uncomfortable standing in front of people and speaking normally. "What if we build an app that emphasized and illustrated the unnerving silence, the fear of the spotlight, and the judging eyes that stare into your soul when you present and then help the users overcome these situations through a voice analyzing system?"
What it does
stage_fright brings the user into the spotlight on a dark stage overlooking an audience of hundreds of expectant people, simulating the common fear performers experience when speaking in front of a crowd. Here in this synthetic environment, the app allows users to practice their presenting ability. It records voice, converts the audio into text, and subsequently analyzes the text for pauses and long words. Finally, it gives helpful feedback to users on how to improve their speaking.
How we built it
Our project consists of two major parts: the VR and the sound/website. The program takes in an audio file and returns how many hesitations, the clarity score, and a graph of the volume used throughout the presentation. These are all visible on the website. Separately, the user can enter a variety of verbal commands to control the VR stage.
Challenges we ran into
Originally, we were going to build an Alexa skill to analyze user input speech-to-text and have Alexa return feedback on it. However, we were unable to get certain voice-related statistics from Alexa, so we switched to IBM to translate speech to text.
Accomplishments that we're proud of
We are proud to say that we've created a program that can help such a wide audience of people throughout the world (including people who suffer from our same fears).
What we learned
Xiao Qing learned the gist of how to make a pretty website and how to develop an Alexa skill (even though it wasn't used in our final project). Emmanuel learned how to use Unity and Occulus, as well as some more C#. Justin brushed up on his Python skills and added IBM to his toolbelt. Emily was the glue that held everything together, making sure that Flask did what it was supposed to do.
What's next for stage_fright
We really want to turn this into an Alexa skill and have Alexa coach the user on improving speech. We also think that improving the VR graphics to make a more realistic environment would be pretty sweet.