New Delhi : An eight-year-old tech enthusiast has developed a voice-controlled artificial intelligence (AI) storytelling device that operates entirely offline using an $8 ESP32-S3 microcontroller, demonstrating the potential of low-cost AI hardware. The system functions without Wi-Fi, cloud services or external APIs, with all AI processing taking place directly on the device.
The AI can generate original short stories, narrate them aloud, display simple animations and respond to voice commands such as “play,” “pause,” “resume” and “next story.” To build the project, Laksh customised the open-source TinyStories LLM for ESP32, originally developed by programmer slvDev. While the base model was designed to create stories offline on the ESP32 chip, Laksh expanded its capabilities by integrating voice interaction, playback controls and animated visuals.
Laksh showed the device generating a humorous story after receiving the voice command, “Make a funny story.” The AI instantly created a new story, read it aloud and displayed animations on the screen. Users could control playback using simple voice commands, highlighting the device’s fully offline functionality. Laksh said the biggest challenge at present is the quality of the text-to-speech (TTS) engine running on the ESP32. Although the AI can generate and narrate stories, the English voice output still sounds basic.
Unlike most children his age, Laksh follows a flexible online education programme that allows him to focus on independent learning, travel and hands-on innovation. He spends much of his time interacting with founders, entrepreneurs and technology professionals, helping him pursue advanced projects from a young age. The project demonstrates how open-source AI models and affordable hardware can be combined to create practical offline AI applications.
Read Also : Rejected Cybersecurity Aspirant Hacks IIT Kanpur Website, Offered Internship Instead