Inspiration
UNSAID was inspired by how often people struggle to express strong or complicated emotions clearly. In emotional moments, words come out rushed, confusing, or harsher than intended, leading to regret and misunderstanding. I wanted a tool that focuses purely on turning emotions into better language, without judgment or advice.
What it does
Translates raw emotional input into clear, respectful language Generates four outputs:
- Clear expression
- Respectful expression
- Detected emotions
- Validation message Improves communication without giving advice or judgment
How I built it
Frontend **: React, Tailwind CSS, Framer Motion **Backend: Node.js, Express Database: MongoDB (session-based storage) AI: Google Gemini with strict safety-first prompt engineering Enforced structured, JSON-only AI responses for predictability
Challenges I ran into
The main challenge was balancing usefulness with safety. Emotional input can easily push AI toward advice or mental health interpretation, so I had to strictly constrain AI behavior. Another challenge was resisting feature creep and keeping the product focused on clarity rather than guidance.
Accomplishments that I'm proud of
- Built a complete, functional full-stack product in a short timeframe
- Designed a safety-first AI system for emotional input
- Created a unique non-clinical approach to emotional communication
- Delivered a calm, accessible, and emotionally aware UI
What I learned
I learned that strong constraints improve both safety and usability. Handling emotional data requires careful design, clear boundaries, and thoughtful UX. I also learned that clarity alone can significantly improve emotional communication.
What's next for UNSAID
Next steps include multi-language support, voice input and output, emotion trend insights, and a developer API. As UNSAID grows, its core mission will remain the same: helping people express emotions clearly and safely.
Log in or sign up for Devpost to join the conversation.