Skip to main content
Official JavaScript/TypeScript SDK for Lokutor AI, providing high-performance, real-time Voice-to-Voice AI.

​
Features

  • ⚡ Ultra-low latency: Optimized WebSocket streaming for real-time interactions.
  • 🗣️ Conversational Intelligence: Integrated STT, LLM, and TTS with VAD and barge-in support.
  • 🌍 Multi-lingual: Support for English, Spanish, French, Portuguese, and Korean.
  • 🎨 Natural Voices: Multiple high-quality male and female voice styles.
  • 💻 TypeScript First: Full type definitions included.

​
Installation

Install the SDK using your favorite package manager: 
npm install @lokutor/sdk

​
Platform Compatibility

The SDK is platform-agnostic and works in both Node.js and Browser environments.

​
For Node.js

To use the voice agent in a terminal or backend service, you’ll need additional libraries for hardware access: 
npm install node-record-lpcm16 speaker
Linux Users: You may need to install ALSA or PulseAudio development headers: 
sudo apt-get install libasound2-dev

​
For Browser

No extra steps are needed. The SDK uses standard WebSocket and Uint8Array types that are natively supported by modern browsers. Use the Web Audio API to capture microphone input and play back the received buffers.