SpeakCheck

Inspiration

From our previous experiences in hackathons, one of the most stressful parts is trying to create the perfect pitch. But after 36 hours of barely sleeping, it can become pretty difficult. Yet, the pitch is crucial to convey the value of your hack. We created SpeakCheck to help hackers, startup founders, and job applicants excel at public speaking.

What it does

SpeakCheck is an AI-powered platform designed to help users improve their public speaking and interview skills. By leveraging advanced video and audio analysis, it provides real-time feedback on both speech delivery and posture, empowering users to communicate more confidently and effectively. It also helps users prepare for their pitches through mock interviews.

How we built it

We built it utilizing the Gemini API for its video and audio understanding capabilities. The backend was developed with Flask for its simplicity and lightweight structure, while the frontend was built using React to create a responsive and interactive user interface.

Challenges we ran into

One of the main challenges we faced was embedding a text-to-speech model to enhance the mock interview experience and make interactions feel more authentic. Integrating this feature in a way that was both seamless and responsive proved to be particularly difficult, especially when syncing it with real-time user feedback.

Accomplishments that we're proud of

Explored new technologies such as the different Gemini abilities and front-end modifications to create an effective user experience and finishing the project.

What we learned

We learned how to utilize Gemini's intensive capabilities, such as video/audio understanding and real-time interaction through effective prompt engineering. We also learned how to utilize Flask efficiently as a backend server and combine it with React for a smooth and responsive user experience.