@andresaya/edge-tts 1.2.4 → 1.2.6

package/README.md

@@ -1,95 +1,91 @@
-# Edge TTS
-
-**Edge TTS** is a powerful Text-to-Speech (TTS) package that leverages Microsoft's Edge capabilities. This package allows you to synthesize speech from text and manage voice options easily through a command-line interface (CLI).
-
-## Features
-
-- **Text-to-Speech**: Convert text into natural-sounding speech using Microsoft Edge's TTS capabilities.
-- **Multiple Voices**: Access a variety of voices to suit your project's needs.
-- **Audio Export Options**: Export synthesized audio in different formats (raw, base64, or directly to a file).
-- **Command-Line Interface**: Use a simple CLI for easy access to functionality.
-- **Easy Integration**: Modular structure allows for easy inclusion in existing projects.
-
-## Installation
-
-You can install Edge TTS via npm. Run the following command in your terminal:
-
-```bash
-bun add @andresaya/edge-tts
-```
-```bash
-npm install @andresaya/edge-tts
-```
-
-
-## Usage
-Command-Line Interface
-
-You can install Edge TTS via npm. bun  Run the following command in your terminal:
-
-```bash
-npm install -g @andresaya/edge-tts
-```
-
-```bash
-bun install --global edge-tts
-```
-
-To synthesize speech from text, use the following command:
-```bash
-edge-tts synthesize -t "Hello, world!" -o hello_world_audio
-```
-
-To list available voices, run:
-
-```bash
-edge-tts voice-list
-```
-
-Integration into Your Project
-To use Edge TTS in your Bun project, you can import it like this:
-
-```js
-import { EdgeTTS } from '@andresaya/edge-tts';
-
-// Initialize the EdgeTTS service
-const tts = new EdgeTTS();
-
-// Get voices
-const voices = await tts.getVoices();  
-console.log(voices);  // Display available voices
-
-// Synthesize text with options for voice, rate, volume, and pitch
-await tts.synthesize("Hello, world!", 'en-US-AriaNeural', {
-    rate: '0%',       // Speech rate (range: -100% to 100%)
-    volume: '0%',     // Speech volume (range: -100% to 100%)
-    pitch: '0Hz'      // Voice pitch (range: -100Hz to 100Hz)
-});
-
-// Export synthesized audio in different formats
-tts.toBase64();   // Get audio as base64
-await tts.toFile("output_audio");       // Save audio to file
-await tts.toRaw();         // Get raw audio buffer
-```
-
-## Export Options
-After synthesizing speech, you can export the audio in various formats:
-
-- `toBase64`: Returns the audio as a Base64 string.
-- `toFile`: Saves the audio to a specified file (e.g., "output.wav").
-- `toRaw`: Returns the raw audio stream.
-
-## PHP Version
-If you want to use Edge TTS with PHP, you can check out the PHP version of this package, [Edge TTS PHP](https://github.com/andresayac/edge-tts-php)
-
-
-## License
-This project is licensed under the GNU General Public License v3 (GPLv3).
-
-## Acknowledgments
-
-We would like to extend our gratitude to the developers and contributors of the following projects for their inspiration and groundwork:
-
-* https://github.com/rany2/edge-tts/tree/master/examples
-* https://github.com/rany2/edge-tts/blob/master/src/edge_tts/util.py
+# Edge TTS
+
+**Edge TTS** is a powerful Text-to-Speech (TTS) package that leverages Microsoft's Edge capabilities. This package allows you to synthesize speech from text and manage voice options easily through a command-line interface (CLI).
+
+## Features
+
+- **Text-to-Speech**: Convert text into natural-sounding speech using Microsoft Edge's TTS capabilities.
+- **Multiple Voices**: Access a variety of voices to suit your project's needs.
+- **Audio Export Options**: Export synthesized audio in different formats (raw, base64, or directly to a file).
+- **Command-Line Interface**: Use a simple CLI for easy access to functionality.
+- **Easy Integration**: Modular structure allows for easy inclusion in existing projects.
+
+## Installation
+
+You can install Edge TTS via npm. Run the following command in your terminal:
+
+```bash
+bun add @andresaya/edge-tts
+```
+```bash
+npm install @andresaya/edge-tts
+```
+
+
+## Usage
+Command-Line Interface
+
+You can install Edge TTS via npm. bun  Run the following command in your terminal:
+
+```bash
+npm install -g @andresaya/edge-tts
+```
+
+To synthesize speech from text, use the following command:
+```bash
+edge-tts synthesize -t "Hello, world!" -o hello_world_audio
+```
+
+To list available voices, run:
+
+```bash
+edge-tts voice-list
+```
+
+Integration into Your Project
+To use Edge TTS in your Bun project, you can import it like this:
+
+```js
+import { EdgeTTS } from '@andresaya/edge-tts';
+
+// Initialize the EdgeTTS service
+const tts = new EdgeTTS();
+
+// Get voices
+const voices = await tts.getVoices();  
+console.log(voices);  // Display available voices
+
+// Synthesize text with options for voice, rate, volume, and pitch
+await tts.synthesize("Hello, world!", 'en-US-AriaNeural', {
+    rate: '0%',       // Speech rate (range: -100% to 100%)
+    volume: '0%',     // Speech volume (range: -100% to 100%)
+    pitch: '0Hz'      // Voice pitch (range: -100Hz to 100Hz)
+});
+
+// Export synthesized audio in different formats
+tts.toBase64();   // Get audio as base64
+await tts.toFile("output_audio");       // Save audio to file
+await tts.toRaw();         // Get raw audio buffer
+```
+
+## Export Options
+After synthesizing speech, you can export the audio in various formats:
+
+- `toBase64`: Returns the audio as a Base64 string.
+- `toFile`: Saves the audio to a specified file (e.g., "output.wav").
+- `toRaw`: Returns the raw audio stream.
+
+## PHP Version
+If you want to use Edge TTS with PHP, you can check out the PHP version of this package, [Edge TTS PHP](https://github.com/andresayac/edge-tts-php)
+
+
+## License
+This project is licensed under the GNU General Public License v3 (GPLv3).
+
+## Acknowledgments
+
+We would like to extend our gratitude to the developers and contributors of the following projects for their inspiration and groundwork:
+
+* https://github.com/rany2/edge-tts/tree/master/examples
+* https://github.com/rany2/edge-tts/blob/master/src/edge_tts/util.py
 * https://github.com/hasscc/hass-edge-tts/blob/main/custom_components/edge_tts/tts.py

package/dist/cli/edge-tts.js

@@ -1,14 +1,12 @@
 #!/usr/bin/env node
-"use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-const SynthesizeCommand_1 = require("../commands/SynthesizeCommand");
-const VoiceListCommand_1 = require("../commands/VoiceListCommand");
-const commander_1 = require("commander");
-const program = new commander_1.Command();
+import { SynthesizeCommand } from '../commands/SynthesizeCommand';
+import { VoiceListCommand } from '../commands/VoiceListCommand';
+import { Command } from 'commander';
+const program = new Command();
 program
-    .version('1.2.0')
-    .addCommand(SynthesizeCommand_1.SynthesizeCommand)
-    .addCommand(VoiceListCommand_1.VoiceListCommand);
+    .version('1.2.6')
+    .addCommand(SynthesizeCommand)
+    .addCommand(VoiceListCommand);
 try {
     program.parse();
 }

package/dist/commands/SynthesizeCommand.js

@@ -1,9 +1,6 @@
-"use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.SynthesizeCommand = void 0;
-const commander_1 = require("commander");
-const EdgeTTS_1 = require("../services/EdgeTTS");
-exports.SynthesizeCommand = new commander_1.Command('synthesize')
+import { Command } from 'commander';
+import { EdgeTTS } from '../services/EdgeTTS';
+export const SynthesizeCommand = new Command('synthesize')
     .description('Edge TTS: synthesize text to audio')
     .option('-t, --text <text>', 'Text to convert to audio')
     .option('-v, --voice [voice]', 'Voice to use for the audio synthesis', 'en-US-AriaNeural')
@@ -17,7 +14,7 @@ exports.SynthesizeCommand = new commander_1.Command('synthesize')
         console.error('Text is required');
         process.exit(1);
     }
-    const tts = new EdgeTTS_1.EdgeTTS();
+    const tts = new EdgeTTS();
     await tts.synthesize(text, voice, { pitch, rate, volume });
     await tts.toFile(`${output}`);
     console.log(`Audio file generated: ${output}.mp3`);

package/dist/commands/VoiceListCommand.js

@@ -1,12 +1,9 @@
-"use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.VoiceListCommand = void 0;
-const commander_1 = require("commander");
-const EdgeTTS_1 = require("../services/EdgeTTS");
-exports.VoiceListCommand = new commander_1.Command('voice-list')
+import { Command } from 'commander';
+import { EdgeTTS } from '../services/EdgeTTS';
+export const VoiceListCommand = new Command('voice-list')
     .description('Get the list of available voices')
     .action(async () => {
-    const tts = new EdgeTTS_1.EdgeTTS();
+    const tts = new EdgeTTS();
     const voices = await tts.getVoices();
     console.log('Lista de voces disponibles:');
     voices.forEach((voice) => {

package/dist/config/constants.js

@@ -1,7 +1,4 @@
-"use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.Constants = void 0;
-exports.Constants = {
+export const Constants = {
     TRUSTED_CLIENT_TOKEN: '6A5AA1D4EAFF4E9FB37E23D68491D6F4',
     WSS_URL: 'wss://speech.platform.bing.com/consumer/speech/synthesize/readaloud/edge/v1',
     VOICES_URL: 'https://speech.platform.bing.com/consumer/speech/synthesize/readaloud/voices/list',

package/dist/index.js

@@ -1,5 +1 @@
-"use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.EdgeTTS = void 0;
-var EdgeTTS_1 = require("./services/EdgeTTS");
-Object.defineProperty(exports, "EdgeTTS", { enumerable: true, get: function () { return EdgeTTS_1.EdgeTTS; } });
+export { EdgeTTS } from './services/EdgeTTS';

package/dist/services/EdgeTTS.js

@@ -1,18 +1,28 @@
-"use strict";
-var __importDefault = (this && this.__importDefault) || function (mod) {
-    return (mod && mod.__esModule) ? mod : { "default": mod };
-};
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.EdgeTTS = void 0;
-const ws_1 = __importDefault(require("ws"));
-const constants_1 = require("../config/constants");
-const fs_1 = require("fs");
-class EdgeTTS {
+import WebSocket, {} from 'ws';
+import { Constants } from '../config/constants';
+import { writeFile } from 'fs/promises';
+import { Buffer } from 'buffer';
+function ensureBuffer(data) {
+    if (Buffer.isBuffer(data)) {
+        return data;
+    }
+    if (data instanceof ArrayBuffer) {
+        return Buffer.from(data);
+    }
+    if (Array.isArray(data)) {
+        return Buffer.concat(data);
+    }
+    if (typeof data === 'string') {
+        return Buffer.from(data, 'utf-8');
+    }
+    throw new Error(`Unsupported RawData type: ${typeof data}`);
+}
+export class EdgeTTS {
     audio_stream = [];
     audio_format = 'mp3';
     ws;
     async getVoices() {
-        const response = await fetch(`${constants_1.Constants.VOICES_URL}?trustedclienttoken=${constants_1.Constants.TRUSTED_CLIENT_TOKEN}`);
+        const response = await fetch(`${Constants.VOICES_URL}?trustedclienttoken=${Constants.TRUSTED_CLIENT_TOKEN}`);
         const data = await response.json();
         return data.map((voice) => {
             delete voice.VoiceTag;
@@ -29,28 +39,56 @@ class EdgeTTS {
         });
     }
     validatePitch(pitch) {
-        if (!/^(-?\d{1,3}Hz)$/.test(pitch)) {
-            throw new Error("Invalid pitch format. Expected format: '-100Hz to 100Hz'.");
+        if (typeof pitch === 'number') {
+            return (pitch >= 0 ? `+${pitch}Hz` : `${pitch}Hz`);
+        }
+        if (!/^[+-]?\d{1,3}(?:\.\d+)?Hz$/.test(pitch)) {
+            throw new Error("Invalid pitch format. Expected format: '-100Hz to +100Hz' or a number.");
         }
         return pitch;
     }
     validateRate(rate) {
-        if (!/^(-?\d{1,3}%)$/.test(rate)) {
-            throw new Error("Invalid rate format. Expected format: '-100% to 100%'.");
+        let rateValue;
+        if (typeof rate === 'string') {
+            rateValue = parseFloat(rate.replace('%', ''));
+            if (isNaN(rateValue))
+                throw new Error("Invalid rate format.");
+        }
+        else {
+            rateValue = rate;
+        }
+        if (rateValue >= 0) {
+            return `+${rateValue}%`;
         }
-        return rate;
+        return `${rateValue}%`;
     }
     validateVolume(volume) {
-        if (!/^(-?\d{1,3}%)$/.test(volume)) {
-            throw new Error("Invalid volume format. Expected format: '-100% to 100%'.");
+        let volumeValue;
+        if (typeof volume === 'string') {
+            volumeValue = parseInt(volume.replace('%', ''), 10);
+            if (isNaN(volumeValue))
+                throw new Error("Invalid volume format.");
+        }
+        else {
+            volumeValue = volume;
         }
-        return volume;
+        if (volumeValue < -100 || volumeValue > 100) {
+            throw new Error("Volume cannot be negative. Expected a value from -100% to 100% (or more).");
+        }
+        return `${volumeValue}%`;
     }
     async synthesize(text, voice = 'en-US-AnaNeural', options = {}) {
         return new Promise((resolve, reject) => {
+            this.audio_stream = [];
             const req_id = this.generateUUID();
-            this.ws = new ws_1.default(`${constants_1.Constants.WSS_URL}?trustedclienttoken=${constants_1.Constants.TRUSTED_CLIENT_TOKEN}&ConnectionId=${req_id}`);
+            this.ws = new WebSocket(`${Constants.WSS_URL}?trustedclienttoken=${Constants.TRUSTED_CLIENT_TOKEN}&ConnectionId=${req_id}`);
             const SSML_text = this.getSSML(text, voice, options);
+            const timeout = setTimeout(() => {
+                if (this.ws && this.ws.readyState === WebSocket.OPEN) {
+                    this.ws.close();
+                }
+                reject(new Error("Synthesis timeout"));
+            }, 30000);
             this.ws.on('open', () => {
                 const message = this.buildTTSConfigMessage();
                 this.ws.send(message);
@@ -60,19 +98,26 @@ class EdgeTTS {
             this.ws.on('message', (data) => {
                 this.processAudioData(data);
             });
-            this.ws.on('close', () => {
-                resolve();
-            });
             this.ws.on('error', (err) => {
+                clearTimeout(timeout);
+                if (this.ws && this.ws.readyState === WebSocket.OPEN) {
+                    this.ws.close();
+                }
                 reject(err);
             });
+            this.ws.on('close', () => {
+                clearTimeout(timeout);
+                resolve();
+            });
         });
     }
     getSSML(text, voice, options = {}) {
-        options.pitch = options.pitch.replace('hz', 'Hz');
-        const pitch = this.validatePitch(options.pitch || '0Hz');
-        const rate = this.validateRate(options.rate || '0%');
-        const volume = this.validateVolume(options.volume || '0%');
+        if (typeof options.pitch === 'string') {
+            options.pitch = options.pitch.replace('hz', 'Hz');
+        }
+        const pitch = this.validatePitch(options.pitch ?? 0);
+        const rate = this.validateRate(options.rate ?? 0);
+        const volume = this.validateVolume(options.volume ?? 0);
         return `<speak version='1.0' xml:lang='en-US'><voice name='${voice}'><prosody pitch='${pitch}' rate='${rate}' volume='${volume}'>${text}</prosody></voice></speak>`;
     }
     buildTTSConfigMessage() {
@@ -80,39 +125,33 @@ class EdgeTTS {
             `{"context":{"synthesis":{"audio":{"metadataoptions":{"sentenceBoundaryEnabled":false,"wordBoundaryEnabled":true},"outputFormat":"audio-24khz-48kbitrate-mono-mp3"}}}}`;
     }
     processAudioData(data) {
-        if (Buffer.isBuffer(data)) {
-            const needle = Buffer.from("Path:audio\r\n");
-            const uint8Data = new Uint8Array(data);
-            const start_ind = uint8Data.indexOf(needle[0]);
-            if (start_ind !== -1) {
-                const audioData = data.subarray(start_ind + needle.length);
-                this.audio_stream.push(audioData);
-            }
-            if (data.includes("Path:turn.end")) {
-                this.ws.close();
-            }
+        const buffer = ensureBuffer(data);
+        const needle = Buffer.from("Path:audio\r\n");
+        const audioStartIndex = buffer.indexOf(new Uint8Array(needle));
+        if (audioStartIndex !== -1) {
+            const audioChunk = buffer.subarray(audioStartIndex + needle.length);
+            this.audio_stream.push(new Uint8Array(audioChunk));
+        }
+        if (buffer.toString().includes("Path:turn.end")) {
+            this.ws?.close();
         }
     }
     async toFile(outputPath) {
-        if (this.audio_stream.length) {
-            const audioBuffer = Buffer.concat(this.audio_stream);
-            const uint8Array = new Uint8Array(audioBuffer);
-            (0, fs_1.writeFileSync)(`${outputPath}.${this.audio_format}`, uint8Array);
-        }
-        else {
-            throw new Error("No audio data available to save.");
-        }
+        const audioBuffer = this.toBuffer();
+        const finalPath = `${outputPath}.${this.audio_format}`;
+        await writeFile(finalPath, new Uint8Array(audioBuffer));
+        return finalPath;
     }
     toRaw() {
-        if (this.audio_stream.length === 0) {
-            throw new Error("No audio data available.");
-        }
-        const audioBuffer = Buffer.concat(this.audio_stream);
-        return audioBuffer.toString('base64');
+        return this.toBase64();
     }
     toBase64() {
-        const audioBuffer = Buffer.concat(this.audio_stream);
-        return audioBuffer.toString('base64');
+        return this.toBuffer().toString('base64');
+    }
+    toBuffer() {
+        if (this.audio_stream.length === 0) {
+            throw new Error("No audio data available. Did you run synthesize() first?");
+        }
+        return Buffer.concat(this.audio_stream);
     }
 }
-exports.EdgeTTS = EdgeTTS;

package/package.json

@@ -1,47 +1,48 @@
-{
-  "name": "@andresaya/edge-tts",
-  "version": "1.2.4",
-  "description": "Edge TTS is a package that allows access to the online text-to-speech service used by Microsoft Edge without the need for Microsoft Edge, Windows, or an API key.",
-  "main": "dist/index.js",
-  "keywords": [
-    "edge",
-    "tts",
-    "text-to-speech",
-    "microsoft",
-    "microsoft edge",
-    "speech-synthesis"
-  ],
-  "author": "Andres Aya",
-  "license": "GPL-3.0-only",
-  "homepage": "https://github.com/andresayac/edge-tts",
-  "bugs": {
-    "url": "https://github.com/andresayac/edge-tts/issues"
-  },
-  "scripts": {
-    "build": "tsc",
-    "build:bin": "bun scripts/build.ts"
-  },
-  "devDependencies": {
-    "@types/bun": "latest",
-    "typescript": "^5.6.2"
-  },
-  "dependencies": {
-    "@types/uuid": "^10.0.0",
-    "commander": "^12.1.0",
-    "events": "^3.3.0",
-    "fs": "^0.0.1-security",
-    "ws": "^8.18.0"
-  },
-  "peerDependencies": {
-    "typescript": "^5.0.0"
-  },
-  "files": [
-    "dist"
-  ],
-  "bin": {
-    "edge-tts": "./dist/cli/edge-tts.js"
-  },
-  "engines": {
-    "bun": ">=0.5.0"
-  }
-}
+{
+  "name": "@andresaya/edge-tts",
+  "version": "1.2.6",
+  "description": "Edge TTS is a package that allows access to the online text-to-speech service used by Microsoft Edge without the need for Microsoft Edge, Windows, or an API key.",
+  "main": "dist/index.js",
+  "keywords": [
+    "edge",
+    "tts",
+    "text-to-speech",
+    "microsoft",
+    "microsoft edge",
+    "speech-synthesis"
+  ],
+  "author": "Andres Aya",
+  "license": "GPL-3.0-only",
+  "homepage": "https://github.com/andresayac/edge-tts",
+  "bugs": {
+    "url": "https://github.com/andresayac/edge-tts/issues"
+  },
+  "scripts": {
+    "build": "tsc",
+    "build:bin": "bun scripts/build.ts"
+  },
+  "devDependencies": {
+    "@types/bun": "latest",
+    "@types/ws": "^8.18.1",
+    "typescript": "^5.6.2"
+  },
+  "dependencies": {
+    "@types/uuid": "^10.0.0",
+    "commander": "^12.1.0",
+    "events": "^3.3.0",
+    "fs": "^0.0.1-security",
+    "ws": "^8.18.0"
+  },
+  "peerDependencies": {
+    "typescript": "^5.0.0"
+  },
+  "files": [
+    "dist"
+  ],
+  "bin": {
+    "edge-tts": "./dist/cli/edge-tts.js"
+  },
+  "engines": {
+    "bun": ">=0.5.0"
+  }
+}

package/dist/cli/edge-tts.d.ts

@@ -1,2 +0,0 @@
-#!/usr/bin/env node
-export {};

package/dist/commands/SynthesizeCommand.d.ts

@@ -1,2 +0,0 @@
-import { Command } from 'commander';
-export declare const SynthesizeCommand: Command;

package/dist/commands/VoiceListCommand.d.ts

@@ -1,2 +0,0 @@
-import { Command } from 'commander';
-export declare const VoiceListCommand: Command;

package/dist/config/constants.d.ts

@@ -1,5 +0,0 @@
-export declare const Constants: {
-    TRUSTED_CLIENT_TOKEN: string;
-    WSS_URL: string;
-    VOICES_URL: string;
-};

package/dist/index.d.ts

@@ -1 +0,0 @@
-export { EdgeTTS } from './services/EdgeTTS';

package/dist/services/EdgeTTS.d.ts

@@ -1,17 +0,0 @@
-export declare class EdgeTTS {
-    private audio_stream;
-    private audio_format;
-    private ws;
-    getVoices(): Promise<any[]>;
-    private generateUUID;
-    private validatePitch;
-    private validateRate;
-    private validateVolume;
-    synthesize(text: string, voice?: string, options?: any): Promise<void>;
-    private getSSML;
-    private buildTTSConfigMessage;
-    private processAudioData;
-    toFile(outputPath: string): Promise<void>;
-    toRaw(): string;
-    toBase64(): string;
-}