Inspiration
Remino Verse is born from a personal need we all share: the desire to hold onto the voices, personalities, and moments of those who’ve shaped us. While many inspirations shaped this journey, one recent parallel we deeply resonated with was N.A.T.A.L.I.E. — the AI created by Riri Williams in Marvel’s Ironheart. In that story, Riri revives the memory and emotional depth of her late best friend. Though our project began weeks before the series aired, the parallels affirmed the emotional and technical direction we were pursuing: AI not just as a tool, but as a companion that can carry memory, warmth, and continuity.
What it does
Remino Verse is a revolutionary AI-powered platform that allows you to create and interact with memory avatars of people meaningful to you. You can bring to life AI representations of loved ones, your inner child, fictional characters, or historical figures. The platform facilitates real conversations through both text and real-time voice, where the AI remembers conversation history and builds relationship context. It's designed as a safe space for connection, memory preservation, and emotional growth, offering a unique way to engage with the past and present.
How we built it
Remino Verse is built on a robust and modern technology stack to deliver a seamless and emotionally resonant experience:
🖥️ Frontend
- React 18 with TypeScript, powered by Vite for fast builds and development.
- Tailwind CSS with Material Design 3 tokens for a responsive, beautiful UI.
- Framer Motion for smooth animations.
- React Router for seamless client-side navigation.
🗄️ Backend
- Supabase provides:
- A secure and scalable PostgreSQL database
- User authentication
- File storage for avatar photos and memory uploads
🤖 AI Integration
- Google Gemini 2.0 Flash drives the conversational AI, enabling contextual and emotionally aware dialogue.
- ElevenLabs handles realistic and configurable voice synthesis (gender, tone, voice ID).
- Web Speech API enables real-time speech recognition for natural voice interactions.
🚀 Deployment
- Deployed and optimized on Netlify for automatic builds, fast delivery, and reliable hosting.
Coding/Developer
- Bolt.new
Challenges we ran into
Our primary challenge has been to make the conversations as natural and human-like as possible, ensuring the AI maintains memory and context across interactions. This involves intricate prompt engineering and careful management of conversational flow to prevent the AI from sounding generic or forgetting past discussions. Achieving a truly seamless and emotionally intelligent dialogue, where the AI feels like a genuine friend, is a continuous endeavor.
Accomplishments that we're proud of
We are incredibly proud of achieving conversations that genuinely feel like talking to a friend. The integration of ElevenLabs has been instrumental in delivering highly realistic and natural voice responses, significantly enhancing the immersive experience. We've successfully built a platform that fosters emotional connection and memory preservation, providing a unique and meaningful way for users to interact with their cherished memories. The real-time voice call functionality with interruption handling is a testament to our technical capabilities.
What we learned
Building Remino Verse has been a profound learning experience. We've gained deep insights into the complexities of creating AI that can maintain long-term memory and consistent personality traits, especially when dealing with sensitive emotional contexts. The intricacies of real-time multi-modal AI interactions, from accurate speech recognition to natural voice synthesis, have pushed our technical boundaries. We've also learned the critical importance of a robust and scalable backend infrastructure, like Supabase, to support complex data structures and ensure a reliable user experience in such a personal application.
What's next for Remino Verse
Our next major goal for Remino Verse is to implement video calls with lifelike avatars. While this feature presents significant technical challenges and is still in development, we are committed to bringing face-to-face interactions to the platform. This will further deepen the emotional connection, allowing users to see the facial expressions and presence of their memory avatars. Beyond video calls, our roadmap includes premium subscription plans, enhanced memory extraction capabilities, a dedicated mobile app, advanced voice cloning, and the potential for group conversations.
Log in or sign up for Devpost to join the conversation.