Agrasya Voice SDK

A simple, secure SDK for integrated voice AI.

Installation

npm install agrasya-voice-sdk

Usage

import { AgrasyaVoiceSDK } from "agrasya-voice-sdk";

// Initialize the SDK
const sdk = new AgrasyaVoiceSDK({
  apiKey: "your-api-key",
  baseUrl: "https://api.agrasya.com",
  language: "en",
  debug: false,
});

// Start a voice session
await sdk.start();

// Handle events
const events = {
  onStatusChange: (status) => console.log("Status:", status),
  onMessageReceived: (message) => console.log("Message:", message),
};

const sdkWithEvents = new AgrasyaVoiceSDK({ apiKey: "your-api-key" }, events);

Methods

Constructor

constructor(config: AgrasyaSDKConfig, events: AgrasyaEvents = {});

Initializes the SDK with the provided configuration and optional event handlers.

start()

async start(): Promise<void>

Starts a voice session by establishing a connection with the backend and initializing WebRTC components.

stop()

async stop(): Promise<void>

Stops the voice session, closes the WebRTC connection, and cleans up resources.

sendMessage()

sendMessage(message: string): void

Sends a message through the data channel.

onStatusChange

onStatusChange(status: "idle" | "initializing" | "connecting" | "connected" | "error"): void

Event handler for status changes.

onMessageReceived

onMessageReceived(message: string): void

Event handler for incoming messages.

Development

# Install dependencies
npm install

# Build the SDK
npm run build

# Watch mode for development
npm run dev

# Run tests
npm test

Publishing

See PUBLISHING.md for instructions on publishing to npm.

License

MIT