agrasya-voice-sdk 1.0.0

package/README.md

@@ -0,0 +1,61 @@
+# Agrasya Voice SDK
+
+A simple, secure SDK for integrated voice AI.
+
+## Installation
+
+```bash
+npm install @agrasya/voice-sdk
+```
+
+## Usage
+
+```typescript
+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);
+```
+
+## Development
+
+```bash
+# Install dependencies
+npm install
+
+# Build the SDK
+npm run build
+
+# Watch mode for development
+npm run dev
+
+# Run tests
+npm test
+
+# Run linting
+npm run lint
+```
+
+## Publishing
+
+See [PUBLISHING.md](./PUBLISHING.md) for instructions on publishing to npm.
+
+## License
+
+MIT

package/dist/index.d.mts

@@ -0,0 +1,62 @@
+/**
+ * Configuration for the Agrasya Voice SDK.
+ */
+interface AgrasyaSDKConfig {
+    /** The API key provided by AGRASYA */
+    apiKey: string;
+    /** The language for the voice session (e.g., 'en', 'hi') */
+    language?: string;
+    /** Custom backend URL (optional) */
+    baseUrl?: string;
+    /** Debug mode */
+    debug?: boolean;
+}
+/**
+ * Events emitted by the Agrasya SDK.
+ */
+interface AgrasyaEvents {
+    /** Emitted when the session starts successfully */
+    onStart?: () => void;
+    /** Emitted when the session ends */
+    onStop?: () => void;
+    /** Emitted when an error occurs */
+    onError?: (error: Error) => void;
+    /** Emitted when a transcript is received */
+    onTranscript?: (data: {
+        text: string;
+        isFinal: boolean;
+    }) => void;
+    /** Emitted when the audio level changes (0-1) */
+    onAudioLevel?: (level: number) => void;
+    /** Emitted when the connection status changes */
+    onStatusChange?: (status: "idle" | "initializing" | "connecting" | "connected" | "error") => void;
+}
+
+/**
+ * AgrasyaVoiceSDK - A simple, secure SDK for integrated voice AI.
+ */
+declare class AgrasyaVoiceSDK {
+    private config;
+    private events;
+    private peerConnection;
+    private dataChannel;
+    private audioStream;
+    private sessionId;
+    private status;
+    constructor(config: AgrasyaSDKConfig, events?: AgrasyaEvents);
+    private log;
+    private setStatus;
+    /**
+     * Starts a voice session.
+     */
+    start(): Promise<void>;
+    /**
+     * Stops the current voice session.
+     */
+    stop(): Promise<void>;
+    private initializeBackendSession;
+    private connectWebRTC;
+    private notifySessionEnd;
+}
+
+export { AgrasyaVoiceSDK };

package/dist/index.d.ts

@@ -0,0 +1,62 @@
+/**
+ * Configuration for the Agrasya Voice SDK.
+ */
+interface AgrasyaSDKConfig {
+    /** The API key provided by AGRASYA */
+    apiKey: string;
+    /** The language for the voice session (e.g., 'en', 'hi') */
+    language?: string;
+    /** Custom backend URL (optional) */
+    baseUrl?: string;
+    /** Debug mode */
+    debug?: boolean;
+}
+/**
+ * Events emitted by the Agrasya SDK.
+ */
+interface AgrasyaEvents {
+    /** Emitted when the session starts successfully */
+    onStart?: () => void;
+    /** Emitted when the session ends */
+    onStop?: () => void;
+    /** Emitted when an error occurs */
+    onError?: (error: Error) => void;
+    /** Emitted when a transcript is received */
+    onTranscript?: (data: {
+        text: string;
+        isFinal: boolean;
+    }) => void;
+    /** Emitted when the audio level changes (0-1) */
+    onAudioLevel?: (level: number) => void;
+    /** Emitted when the connection status changes */
+    onStatusChange?: (status: "idle" | "initializing" | "connecting" | "connected" | "error") => void;
+}
+
+/**
+ * AgrasyaVoiceSDK - A simple, secure SDK for integrated voice AI.
+ */
+declare class AgrasyaVoiceSDK {
+    private config;
+    private events;
+    private peerConnection;
+    private dataChannel;
+    private audioStream;
+    private sessionId;
+    private status;
+    constructor(config: AgrasyaSDKConfig, events?: AgrasyaEvents);
+    private log;
+    private setStatus;
+    /**
+     * Starts a voice session.
+     */
+    start(): Promise<void>;
+    /**
+     * Stops the current voice session.
+     */
+    stop(): Promise<void>;
+    private initializeBackendSession;
+    private connectWebRTC;
+    private notifySessionEnd;
+}
+
+export { AgrasyaVoiceSDK };

package/dist/index.js

@@ -0,0 +1,185 @@
+"use strict";
+var __defProp = Object.defineProperty;
+var __getOwnPropDesc = Object.getOwnPropertyDescriptor;
+var __getOwnPropNames = Object.getOwnPropertyNames;
+var __hasOwnProp = Object.prototype.hasOwnProperty;
+var __export = (target, all) => {
+  for (var name in all)
+    __defProp(target, name, { get: all[name], enumerable: true });
+};
+var __copyProps = (to, from, except, desc) => {
+  if (from && typeof from === "object" || typeof from === "function") {
+    for (let key of __getOwnPropNames(from))
+      if (!__hasOwnProp.call(to, key) && key !== except)
+        __defProp(to, key, { get: () => from[key], enumerable: !(desc = __getOwnPropDesc(from, key)) || desc.enumerable });
+  }
+  return to;
+};
+var __toCommonJS = (mod) => __copyProps(__defProp({}, "__esModule", { value: true }), mod);
+
+// src/index.ts
+var index_exports = {};
+__export(index_exports, {
+  AgrasyaVoiceSDK: () => AgrasyaVoiceSDK
+});
+module.exports = __toCommonJS(index_exports);
+var AgrasyaVoiceSDK = class {
+  config;
+  events;
+  peerConnection = null;
+  dataChannel = null;
+  audioStream = null;
+  sessionId = null;
+  status = "idle";
+  constructor(config, events = {}) {
+    this.config = {
+      baseUrl: "http://localhost:4001",
+      language: "en",
+      debug: false,
+      ...config
+    };
+    this.events = events;
+  }
+  log(...args) {
+    if (this.config.debug) {
+      console.log("[AgrasyaSDK]", ...args);
+    }
+  }
+  setStatus(status) {
+    this.status = status;
+    this.events.onStatusChange?.(status);
+  }
+  /**
+   * Starts a voice session.
+   */
+  async start() {
+    if (this.status !== "idle") {
+      throw new Error("Session is already in progress or initializing");
+    }
+    try {
+      this.setStatus("initializing");
+      this.log("Initializing session...");
+      const sessionData = await this.initializeBackendSession();
+      this.sessionId = sessionData.sessionId;
+      await this.connectWebRTC(sessionData.ephemeralKey);
+      this.setStatus("connected");
+      this.events.onStart?.();
+    } catch (error) {
+      this.setStatus("error");
+      const err = error instanceof Error ? error : new Error(String(error));
+      this.events.onError?.(err);
+      throw err;
+    }
+  }
+  /**
+   * Stops the current voice session.
+   */
+  async stop() {
+    this.log("Stopping session...");
+    if (this.peerConnection) {
+      this.peerConnection.close();
+      this.peerConnection = null;
+    }
+    if (this.dataChannel) {
+      this.dataChannel.close();
+      this.dataChannel = null;
+    }
+    if (this.audioStream) {
+      this.audioStream.getTracks().forEach((track) => track.stop());
+      this.audioStream = null;
+    }
+    if (this.sessionId) {
+      this.notifySessionEnd(this.sessionId).catch(
+        (err) => this.log("Failed to notify session end", err)
+      );
+      this.sessionId = null;
+    }
+    this.setStatus("idle");
+    this.events.onStop?.();
+  }
+  async initializeBackendSession() {
+    const url = `${this.config.baseUrl}/v1/agrasya/client-token?language=${this.config.language}`;
+    const response = await fetch(url, {
+      method: "GET",
+      headers: {
+        "x-agrasya-client-api-key": this.config.apiKey,
+        "Content-Type": "application/json"
+      }
+    });
+    if (!response.ok) {
+      const errorData = await response.json().catch(() => ({ message: response.statusText }));
+      throw new Error(
+        `Failed to initialize session: ${errorData.message || response.statusText}`
+      );
+    }
+    const result = await response.json();
+    if (!result.success) {
+      throw new Error(result.message || "Backend initialization failed");
+    }
+    return result.data;
+  }
+  async connectWebRTC(ephemeralKey) {
+    this.setStatus("connecting");
+    this.log("Connecting to AI Provider via WebRTC...");
+    this.peerConnection = new RTCPeerConnection();
+    this.peerConnection.ontrack = (event) => {
+      this.log("Received remote audio track");
+      const audio = new Audio();
+      audio.srcObject = event.streams[0];
+      audio.play().catch((err) => this.log("Failed to play audio", err));
+    };
+    this.audioStream = await navigator.mediaDevices.getUserMedia({
+      audio: true
+    });
+    this.audioStream.getTracks().forEach((track) => {
+      if (this.peerConnection && this.audioStream) {
+        this.peerConnection.addTrack(track, this.audioStream);
+      }
+    });
+    this.dataChannel = this.peerConnection.createDataChannel("oai-events");
+    this.dataChannel.onmessage = (event) => {
+      const data = JSON.parse(event.data);
+      this.log("Received event:", data.type);
+      if (data.type === "response.audio_transcript.delta") {
+        this.events.onTranscript?.({ text: data.delta, isFinal: false });
+      } else if (data.type === "response.audio_transcript.done") {
+        this.events.onTranscript?.({ text: data.transcript, isFinal: true });
+      }
+    };
+    const offer = await this.peerConnection.createOffer();
+    await this.peerConnection.setLocalDescription(offer);
+    const baseUrl = "https://api.openai.com/v1/realtime";
+    const model = "gpt-4o-realtime-preview-2024-12-17";
+    const sdpResponse = await fetch(`${baseUrl}?model=${model}`, {
+      method: "POST",
+      body: offer.sdp,
+      headers: {
+        Authorization: `Bearer ${ephemeralKey}`,
+        "Content-Type": "application/sdp"
+      }
+    });
+    if (!sdpResponse.ok) {
+      throw new Error(`OpenAI SDP exchange failed: ${sdpResponse.statusText}`);
+    }
+    const answer = {
+      type: "answer",
+      sdp: await sdpResponse.text()
+    };
+    await this.peerConnection.setRemoteDescription(answer);
+    this.log("WebRTC connection established");
+  }
+  async notifySessionEnd(sessionId) {
+    const url = `${this.config.baseUrl}/v1/agrasya/session-end/${sessionId}`;
+    await fetch(url, {
+      method: "PUT",
+      headers: {
+        "x-agrasya-client-api-key": this.config.apiKey,
+        "Content-Type": "application/json"
+      }
+    });
+  }
+};
+// Annotate the CommonJS export names for ESM import in node:
+0 && (module.exports = {
+  AgrasyaVoiceSDK
+});

package/dist/index.mjs

@@ -0,0 +1,160 @@
+// src/index.ts
+var AgrasyaVoiceSDK = class {
+  config;
+  events;
+  peerConnection = null;
+  dataChannel = null;
+  audioStream = null;
+  sessionId = null;
+  status = "idle";
+  constructor(config, events = {}) {
+    this.config = {
+      baseUrl: "http://localhost:4001",
+      language: "en",
+      debug: false,
+      ...config
+    };
+    this.events = events;
+  }
+  log(...args) {
+    if (this.config.debug) {
+      console.log("[AgrasyaSDK]", ...args);
+    }
+  }
+  setStatus(status) {
+    this.status = status;
+    this.events.onStatusChange?.(status);
+  }
+  /**
+   * Starts a voice session.
+   */
+  async start() {
+    if (this.status !== "idle") {
+      throw new Error("Session is already in progress or initializing");
+    }
+    try {
+      this.setStatus("initializing");
+      this.log("Initializing session...");
+      const sessionData = await this.initializeBackendSession();
+      this.sessionId = sessionData.sessionId;
+      await this.connectWebRTC(sessionData.ephemeralKey);
+      this.setStatus("connected");
+      this.events.onStart?.();
+    } catch (error) {
+      this.setStatus("error");
+      const err = error instanceof Error ? error : new Error(String(error));
+      this.events.onError?.(err);
+      throw err;
+    }
+  }
+  /**
+   * Stops the current voice session.
+   */
+  async stop() {
+    this.log("Stopping session...");
+    if (this.peerConnection) {
+      this.peerConnection.close();
+      this.peerConnection = null;
+    }
+    if (this.dataChannel) {
+      this.dataChannel.close();
+      this.dataChannel = null;
+    }
+    if (this.audioStream) {
+      this.audioStream.getTracks().forEach((track) => track.stop());
+      this.audioStream = null;
+    }
+    if (this.sessionId) {
+      this.notifySessionEnd(this.sessionId).catch(
+        (err) => this.log("Failed to notify session end", err)
+      );
+      this.sessionId = null;
+    }
+    this.setStatus("idle");
+    this.events.onStop?.();
+  }
+  async initializeBackendSession() {
+    const url = `${this.config.baseUrl}/v1/agrasya/client-token?language=${this.config.language}`;
+    const response = await fetch(url, {
+      method: "GET",
+      headers: {
+        "x-agrasya-client-api-key": this.config.apiKey,
+        "Content-Type": "application/json"
+      }
+    });
+    if (!response.ok) {
+      const errorData = await response.json().catch(() => ({ message: response.statusText }));
+      throw new Error(
+        `Failed to initialize session: ${errorData.message || response.statusText}`
+      );
+    }
+    const result = await response.json();
+    if (!result.success) {
+      throw new Error(result.message || "Backend initialization failed");
+    }
+    return result.data;
+  }
+  async connectWebRTC(ephemeralKey) {
+    this.setStatus("connecting");
+    this.log("Connecting to AI Provider via WebRTC...");
+    this.peerConnection = new RTCPeerConnection();
+    this.peerConnection.ontrack = (event) => {
+      this.log("Received remote audio track");
+      const audio = new Audio();
+      audio.srcObject = event.streams[0];
+      audio.play().catch((err) => this.log("Failed to play audio", err));
+    };
+    this.audioStream = await navigator.mediaDevices.getUserMedia({
+      audio: true
+    });
+    this.audioStream.getTracks().forEach((track) => {
+      if (this.peerConnection && this.audioStream) {
+        this.peerConnection.addTrack(track, this.audioStream);
+      }
+    });
+    this.dataChannel = this.peerConnection.createDataChannel("oai-events");
+    this.dataChannel.onmessage = (event) => {
+      const data = JSON.parse(event.data);
+      this.log("Received event:", data.type);
+      if (data.type === "response.audio_transcript.delta") {
+        this.events.onTranscript?.({ text: data.delta, isFinal: false });
+      } else if (data.type === "response.audio_transcript.done") {
+        this.events.onTranscript?.({ text: data.transcript, isFinal: true });
+      }
+    };
+    const offer = await this.peerConnection.createOffer();
+    await this.peerConnection.setLocalDescription(offer);
+    const baseUrl = "https://api.openai.com/v1/realtime";
+    const model = "gpt-4o-realtime-preview-2024-12-17";
+    const sdpResponse = await fetch(`${baseUrl}?model=${model}`, {
+      method: "POST",
+      body: offer.sdp,
+      headers: {
+        Authorization: `Bearer ${ephemeralKey}`,
+        "Content-Type": "application/sdp"
+      }
+    });
+    if (!sdpResponse.ok) {
+      throw new Error(`OpenAI SDP exchange failed: ${sdpResponse.statusText}`);
+    }
+    const answer = {
+      type: "answer",
+      sdp: await sdpResponse.text()
+    };
+    await this.peerConnection.setRemoteDescription(answer);
+    this.log("WebRTC connection established");
+  }
+  async notifySessionEnd(sessionId) {
+    const url = `${this.config.baseUrl}/v1/agrasya/session-end/${sessionId}`;
+    await fetch(url, {
+      method: "PUT",
+      headers: {
+        "x-agrasya-client-api-key": this.config.apiKey,
+        "Content-Type": "application/json"
+      }
+    });
+  }
+};
+export {
+  AgrasyaVoiceSDK
+};

package/package.json

@@ -0,0 +1,16 @@
+{
+  "name": "agrasya-voice-sdk",
+  "version": "1.0.0",
+  "main": "dist/index.js",
+  "module": "dist/index.mjs",
+  "types": "dist/index.d.ts",
+  "files": [
+    "dist"
+  ],
+  "scripts": {
+    "build": "tsup src/index.ts --format cjs,esm --dts",
+    "dev": "tsup src/index.ts --format cjs,esm --watch --dts",
+    "lint": "eslint src/**/*.ts",
+    "test": "jest"
+  }
+}