Inspiration
Raíz was inspired by the experiences of immigrant kids who grow up navigating two worlds. Many speak one language at home and another in public, often being told their home language is “broken.” Over time, this creates shame, loss of confidence, and distance from identity. We wanted to build a tool that helps people reconnect with their roots through language, without fear of judgment.
What it does
Raíz is an AI-powered conversational platform that helps users practice their home language through natural dialogue. It focuses on real conversation, phonetics, and tone, allowing users to build confidence and fluency in a safe, supportive environment.
How we built it
We built Raíz as a web application using browser-based speech recognition and text-to-speech for real-time interaction. A backend LLM generates adaptive, human-like responses, while phonetic and tone-aware feedback helps users improve naturally through conversation.
Challenges we ran into
Making conversations feel natural was our biggest challenge. Handling speech interruptions, reducing latency, and avoiding robotic responses required careful design. We also had to balance providing helpful feedback without making users feel corrected or judged.
Accomplishments that we're proud of
We built a working end-to-end conversational system that listens, responds, and speaks naturally. More importantly, we created a product rooted in empathy that addresses a real and often overlooked problem.
What we learned
We learned that language learning is as much emotional as it is technical. Tone, confidence, and safety matter just as much as accuracy. We also gained experience building real-time, voice-driven AI systems.
What's next for Raíz
Next, we plan to add support for more languages, deeper phonetic analysis, personalized learning paths, and community-driven conversation scenarios to help users practice real-life situations.
Log in or sign up for Devpost to join the conversation.