First-Voice.AI

logo
logo
logo
logo

FirstVoice AI: A Scalable, Compassionate Mental Health Assistant

Inspiration

The app was inspired by the urgent need for accessible mental health support in the United States. Over 160 million Americans live in areas with a shortage of mental health professionals, and the current system lacks the capacity to meet this demand. We realized that we could leverage artificial intelligence to help support individuals in need and bridge the gap between society and mental health professionals.

What We Built

FirstVoice AI is an AI-powered mental health support assistant designed to listen, understand, and support users through its empathetic conversation. It serves as a warm, non-judgmental "first voice" for people who are seeking help, but are not yet willing to or have access to a licensed professional. The app includes real-time emotional state detection based on a user's responses to our prompts. We create summaries, suggestions, and resources based off our conversations and provide them in a readily-made PDF document for the user to refer to in a single, clear place.

How We Built It

We used a combination of: Natural Language Processing (NLP) models for emotional tone analysis and summarization. OpenAI's language models for empathy-driven conversations. We used secure APIs to fetch local mental health resources and hotlines based on the user’s location. Supabase as our backend to manage user authentication, store structured conversation summaries, and log interaction data securely

What We Learned

Crafting AI that feels genuinely empathetic takes more than good language—it takes careful tuning and iteration.Real-time NLP requires thoughtful architecture to ensure responsiveness and emotional intelligence. Security and privacy are critical when dealing with sensitive mental health information—we prioritized this from the start.

Challenges We Faced

Balancing speed with depth: Building a fast yet thoughtful conversation model required careful optimization. Managing sensitive topics without overstepping AI’s limits—we had to ensure the assistant recognized when to escalate to human help. Integrating multiple APIs and tools (NLP, vector search, geolocation) into one smooth user experience.

What's Next for FirstVoice AI

We plan to improve the app's personalization through suggesting specific therapists, psychiatrists, or mental health centers close to a user based on their location. We also wish to improve the application's accessibility by implementing support for multiple languages.