BThere

Inspiration

BThere emerged from a genuine desire to strengthen friendships by addressing the subtle challenges in understanding friends on a deeper level. The team recognized that nuanced conversations often go unnoticed, hindering meaningful support and genuine interactions. In the context of the COVID-19 pandemic, the shift to virtual communication intensified these challenges, making it harder to connect on a profound level. Lockdowns and social distancing amplified feelings of isolation, and the absence of in-person cues made understanding friends even more complex. BThere aims to use advanced technologies to overcome these obstacles, fostering stronger and more authentic connections in a world where the value of meaningful interactions has become increasingly apparent.

What it does

BThere is a friend-assisting application that utilizes cutting-edge technologies to analyze conversations and provide insightful suggestions for users to connect with their friends on a deeper level. By recording conversations through video, the application employs Google Cloud's facial recognition and speech-to-text APIs to understand the friend's mood, likes, and dislikes. The OpenAI API generates personalized suggestions based on this analysis, offering recommendations to uplift a friend in moments of sadness or providing conversation topics and activities for neutral or happy states. The backend, powered by Python Flask, handles data storage using Firebase for authentication and data persistence. The frontend is developed using React, JavaScript, Next.js, HTML, and CSS, creating a user-friendly interface for seamless interaction.

How we built it

BThere involves a multi-faceted approach, incorporating various technologies and platforms to achieve its goals. The recording feature utilizes WebRTC for live video streaming to the backend through sockets, but also allows users to upload videos for analysis. Google Cloud's facial recognition API identifies facial expressions, while the speech-to-text API extracts spoken content. The combination of these outputs serves as input for the OpenAI API, generating personalized suggestions. The backend, implemented in Python Flask, manages data storage in Firebase, ensuring secure authentication and persistent data access. The frontend, developed using React, JavaScript, Next.js, HTML, and CSS, delivers an intuitive user interface.

Accomplishments that we're proud of

Successfully integrating multiple technologies into a cohesive and functional application
Developing a user-friendly frontend for a seamless experience
Implementing real-time video streaming using WebRTC and sockets
Leveraging Google Cloud and OpenAI APIs for advanced facial recognition, speech-to-text, and suggestion generation

What's next for BThere

Continuously optimizing the speech to text and emotion analysis model for improved accuracy with different accents, speech mannerisms, and languages
Exploring advanced natural language processing (NLP) techniques to enhance conversational analysis
Enhancing user experience through further personalization and more privacy features
Conducting user feedback sessions to refine and expand the application's capabilities