Heartbridge

Your only lifetime companion to mitigate loneliness and depression

Comment

Inspiration

The increasing issues of loneliness and isolation in today's fast-paced world inspired us to create Heartbridge. Many individuals, especially those living alone or going through tough times, lack meaningful conversations. We envisioned an AI-driven platform that could provide continuous companionship, support, and positivity through engaging conversations, helping users combat feelings of loneliness.

What it does

Heartbridge is an AI-powered application that generates endless, meaningful conversations between two virtual personas. These personas can converse on various topics, and the user can join in at any time, either through voice or by typing their inputs. The aim is to provide a comforting, non-judgmental space where users can feel less isolated and more connected through the power of conversation.

How we built it

We developed Heartbridge using Flask for the web application, Google Generative AI (Gemini) to generate dynamic and engaging dialogue, and gTTS (Google Text-to-Speech) to convert the generated conversations into speech. The application also incorporates speech recognition for users to interact with the AI using their voice. We integrated the project into a web interface using HTML/CSS and JavaScript for smooth user interaction, while threading helps keep the conversation continuous.

Challenges we ran into

We faced several technical challenges throughout the project. Initially, we encountered issues with pyttsx3, which was replaced with gTTS for more efficient text-to-speech generation. We also dealt with path issues when saving audio files and microphone support problems while deploying the project on PythonAnywhere. Additionally, integrating smooth voice input with dynamic conversation generation was a complex challenge that we overcame through thorough testing and debugging.

Accomplishments that we're proud of

We are proud of building an application that can provide continuous AI-generated conversation, offering comfort to those who feel lonely. The successful integration of real-time voice input and output, coupled with dynamic conversational flow, was a significant achievement. Our biggest accomplishment is creating a solution that can truly make a positive impact on mental health.

What we learned

Throughout the development of Heartbridge, we gained valuable insights into working with Google Generative AI, voice recognition tools, and handling real-time audio playback in web applications. We also learned the importance of tackling real-world issues like loneliness using AI and creating technology with emotional impact in mind.

What's next for Heartbridge

We plan to expand Heartbridge's capabilities by incorporating more diverse conversational topics and emotional intelligence features. We aim to improve the application's ability to detect a user's emotional state and adjust responses accordingly, making conversations even more personalized and supportive. Additionally, we are exploring deploying the project on more platforms to make it accessible to a wider audience.

Built With

Share this project:

Updates