Let's Present

Here's a screenshot of our speedometer for the user!

Inspiration

"Of all the phobias out there, public speaking is considered the highest." (brandongaille.org). As a result, the presentation is influenced by negative factors, which includes but is not limited to, filler words, speaking too fast, fidgeting, and nervousness. We wanted to combat this by helping the user of our website prepare for the best speech possible!

What it does

To combat this fear, we created a website that includes a speedometer that measures how fast the user is speaking during a presentation. The average "normal" speed for a presentation is about 100-150 words per minute. When the speaker goes over this average, the speedometer will then switch to too fast and prompt the speaker to slow down. Our website also senses when the user displays nervousness or fear and outputs motivational quotes to the corner of the screen to encourage the user to embrace confidence!

How I built it

We coded our application in Python. We used Google's speech-to-text api to translate a live speech to text which was then broken by each word with a timestamp to determine the amount of time it took for the user to say each word. The speech data was then stored into MongoDB so that analysis could be run. We then created several functions to get this data and we used math to determine the user's words per minute. The words per minute was then hosted on an api using FastApi that the website then accessed to display on the speedometer for the user. We also used Google's vision api to track the facial expressions of the user. We then collected the joy_likelihood of the user, which was on a level from 1 to 5 (5 being the best). If the user's level of joy was below 3, then we created a function to display a motivational quote.

Challenges I ran into

None of our team members knew how to use MongoDB or create an api so that was a huge challenge for us. We also never made use of Google APIs to work so it was a huge rollercoaster trying to debug all the errors we received (for example: getting specific aspects of the JSON file).. We also had a lot of trouble integrating the front-end and back-end.

Accomplishments that I'm proud of

We are really proud that we were finally able to link the back-end and front-end. Writing code to integrate google API with information received from the website and store it into the mongo database to later analyze and then output back onto the website was a huge challenge. To finally see that work was very relieving.

What I learned

We learned so much from this project. We learned how to utilize apis, how to use mongodb for the first time, how to make an api using fastapi, and how to properly integrate front-end and back-end!