solace

Home Screen
Lil Guy

Inspiration Solace was inspired by the cyclical nature of the sun—bright, yet always going through phases. Just like the sun, our emotional states shift over time, but wellness and gratitude can be practiced at every stage. With Solace, we aimed to create a companion that gently supports users through these phases, encouraging a proactive approach to mental wellness.

What it does Solace is an emotional wellness companion that uses real-time emotion detection to offer personalized tips based on the user’s mood. By analyzing facial expressions and emotional cues, Solace provides relevant, uplifting suggestions to help users feel centered and supported, whether they’re having a good day or facing challenges.

How we built it We developed Solace using a variety of tools and technologies to create a seamless user experience:

Frontend Design: Built using HTML, CSS, and JavaScript for a responsive, visually appealing interface. Design & Prototyping: We used Figma and Anima to design and refine Solace’s interface, ensuring a user-centered approach. Backend: Built with Python Flask to handle routing, authentication, and server-side logic. Authentication: Integrated Auth0 for secure, user-friendly authentication, making login and logout smooth and secure. Emotion Detection: Leveraged the FER (Facial Expression Recognition) Python library to capture and analyze facial expressions, identifying emotions such as happiness, sadness, and more. Natural Language Processing: Used a Hugging Face model to generate personalized wellness tips, adapting each tip to the user's emotional context. Challenges we ran into One of the biggest challenges was choosing the right model for real-time emotion analysis while ensuring compatibility with our web-based setup. Integrating the FER Python library with Flask presented technical hurdles, as did fine-tuning the Hugging Face model to generate meaningful, non-generic tips. Balancing security, ease of access, and the overall user experience with Auth0 also required careful consideration.

Accomplishments that we're proud of We’re proud of creating a comprehensive emotional wellness tool that combines real-time emotion detection with personalized support. Integrating various technologies—FER, Hugging Face, and Auth0—into a smooth user experience is something we’re particularly proud of, as it brings both practicality and empathy to the application.

What we learned Through Solace, we learned how to integrate various libraries, authentication systems, and design tools to create a cohesive, impactful project. Working with FER and Hugging Face reinforced our understanding of machine learning in wellness applications, while Auth0 deepened our skills in implementing secure, user-friendly authentication.

What we struggled with It was incredibly more difficult than expected to deploy the web app on a server for the world to see. We initially got a .tech domain and everything, but it ended up being a big hassle with lots of free trials, out-of-memory errors, and so much more. In the end, the time constraints limited this to a locally hosted situation.

What's next for Solace Next, we plan to enhance Solace’s emotion detection and tip generation, potentially integrating more advanced models to improve accuracy and personalization. We’re also considering adding features like gratitude journaling, positivity prompts, and guided mindfulness exercises, aiming to make Solace a holistic wellness companion for all phases of life.