Skip to main content
Build AI Voice Agent servers with the Kuralit Python SDK. Create conversational AI systems that can listen, think, and act in real-time.

​
What You Can Build

With the Python SDK, you can build:
  • AI Voice Agent Servers - WebSocket servers that process voice input and respond with AI
  • Customer Support AI Voice Agents - Handle support calls with tools for tickets and FAQs
  • Voice Assistant AI Voice Agents - Personal assistants with calendar, weather, and reminders
  • Enterprise AI Voice Agents - Business applications with API integrations

​
Key Features

🤖 AI Agents

Build conversational AI agents with context and personality

🛠️ Tools & Function Calling

Add capabilities with Python functions and REST API tools

🎤 Real-Time Voice

Process voice input with STT, VAD, and turn detection

​
Quick Start

Build your first AI Voice Agent server in 5 minutes: 
from kuralit.server.agent_session import AgentSession
from kuralit.server.websocket_server import create_app

# Create your AI Voice Agent
agent_session = AgentSession(
    stt="deepgram/nova-2:en-US",      # 🎤 Real-Time Voice
    llm="gemini/gemini-2.0-flash-001", # 🤖 AI Agent
    vad="silero/v3",
    turn_detection="multilingual/v1",
    instructions="You are a helpful AI Voice Agent assistant"
)

# Create WebSocket server
app = create_app(
    api_key_validator=lambda key: key == "demo-api-key",
    agent_session=agent_session,
)

# Run server
import uvicorn
uvicorn.run(app, host="0.0.0.0", port=8000)
Quickstart →
Monitoring: Use the Kuralit Dashboard to monitor your server in real-time, track metrics, and debug conversations.

​
Core Concepts

​
AI Voice Agents

AI Voice Agents are the core of your application. They combine:
  • AI Agents - Conversational intelligence with LLMs
  • Tools - Function calling for capabilities
  • Voice - Real-time speech-to-text and turn detection
Learn about building AI Voice Agents →

​
Plugins

Kuralit uses a plugin-based architecture. Choose from:
  • STT Plugins: Deepgram, Google Cloud STT
  • LLM Plugins: Gemini
  • VAD Plugins: Silero
  • Turn Detection: Multilingual
Explore integrations →

​
Tools

Add capabilities to your AI Voice Agents with:
  • Custom Python Functions - Write your own tools
  • REST API Tools - Load from Postman collections
  • Toolkits - Group related tools together
Learn about tools →

​
Documentation Structure

Basics

Core concepts: Agents, Tools, Voice, Sessions

Plugins

STT, LLM, VAD, and turn detection plugins

API Reference

Complete API documentation

Examples

Working code examples

​
Installation

pip install kuralit
Full installation guide →

​
Next Steps

  1. Quickstart - Connect your existing API to an AI Voice Agent
  2. Learn about Agents - Deep dive into building AI Voice Agents
  3. Add Tools - Make your agent capable
  4. Configure Voice - Set up real-time voice streaming