ElatoAI

Active
GitHub TypeScript NOASSERTION

Description

Realtime Voice AI on Arduino ESP32 with 100+ Voice AI Models for AI Toys, Companions, and Devices. Supports OpenAI Realtime, Gemini, Grok, and Eleven Labs.

Key Features

  • Realtime speech-to-speech AI powered by OpenAI, Gemini, Grok, Eleven Labs, and Hume AI on ESP32
  • 100+ voice AI models supported via Cloudflare Workers and Deno Edge Functions
  • Secure WebSocket communication with Opus audio compression for high-quality streaming
  • Server-side VAD turn detection for natural conversation flow management
  • OTA firmware updates and WiFi management with captive portal on ESP32 devices
  • Custom AI agent creation with personality, voice customization, and pitch factor control

Use Cases

💡 Build smart AI toys and companions with real-time voice conversation on ESP32 hardware
💡 Create custom voice AI agents with different personalities for IoT devices
💡 Deploy global low-latency voice AI pipelines using Cloudflare Edge Functions
💡 Prototype voice-enabled embedded products with OpenAI Realtime API integration
💡 Develop WebRTC and WebSocket-based voice interfaces for web and IoT applications

Tags

voice agent typescript embedded llm gemini

Categories

⚡ Agent Tools

Quick Start

Clone the repo, open the project in PlatformIO or Arduino IDE, configure your ESP32 board settings, flash the firmware, then use the ElatoAI webapp to register your device and connect to a voice AI provider like OpenAI Realtime API.
Visit GitHub

Related Projects