@andresaya/edge-tts 1.3.0 → 1.3.2

package/README.md

@@ -1,4 +1,5 @@
 # Edge TTS
+[![Ask DeepWiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/andresayac/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).
 

package/dist/cli/edge-tts.js

@@ -6,7 +6,7 @@ const VoiceListCommand_1 = require("../commands/VoiceListCommand");
 const commander_1 = require("commander");
 const program = new commander_1.Command();
 program
-    .version('1.2.7')
+    .version('1.3.2')
     .addCommand(SynthesizeCommand_1.SynthesizeCommand)
     .addCommand(VoiceListCommand_1.VoiceListCommand);
 try {

package/dist/config/constants.d.ts

@@ -2,4 +2,5 @@ export declare const Constants: {
     TRUSTED_CLIENT_TOKEN: string;
     WSS_URL: string;
     VOICES_URL: string;
+    USER_AGENT: string;
 };

package/dist/config/constants.js

@@ -5,4 +5,5 @@ exports.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',
+    USER_AGENT: 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/140.0.0.0 Safari/537.36 Edg/140.0.0.0'
 };

package/dist/config/formats.d.ts

@@ -0,0 +1,81 @@
+export declare const AUDIO_FORMATS_STREAM: {
+    readonly 'raw-8khz-8bit-mono-alaw': "audio/basic";
+    readonly 'raw-8khz-8bit-mono-mulaw': "audio/basic";
+    readonly 'raw-8khz-16bit-mono-pcm': "audio/basic";
+    readonly 'raw-16khz-16bit-mono-pcm': "audio/basic";
+    readonly 'raw-22050hz-16bit-mono-pcm': "audio/basic";
+    readonly 'raw-24khz-16bit-mono-pcm': "audio/basic";
+    readonly 'raw-44100hz-16bit-mono-pcm': "audio/basic";
+    readonly 'raw-48khz-16bit-mono-pcm': "audio/basic";
+    readonly 'raw-16khz-16bit-mono-truesilk': "audio/SILK";
+    readonly 'raw-24khz-16bit-mono-truesilk': "audio/SILK";
+    readonly 'audio-16khz-32kbitrate-mono-mp3': "audio/mpeg";
+    readonly 'audio-16khz-64kbitrate-mono-mp3': "audio/mpeg";
+    readonly 'audio-16khz-128kbitrate-mono-mp3': "audio/mpeg";
+    readonly 'audio-24khz-48kbitrate-mono-mp3': "audio/mpeg";
+    readonly 'audio-24khz-96kbitrate-mono-mp3': "audio/mpeg";
+    readonly 'audio-24khz-160kbitrate-mono-mp3': "audio/mpeg";
+    readonly 'audio-48khz-96kbitrate-mono-mp3': "audio/mpeg";
+    readonly 'audio-48khz-192kbitrate-mono-mp3': "audio/mpeg";
+    readonly 'audio-16khz-16bit-32kbps-mono-opus': "audio/opus";
+    readonly 'audio-24khz-16bit-24kbps-mono-opus': "audio/opus";
+    readonly 'audio-24khz-16bit-48kbps-mono-opus': "audio/opus";
+    readonly 'ogg-16khz-16bit-mono-opus': "audio/ogg; codecs=opus; rate=16000";
+    readonly 'ogg-24khz-16bit-mono-opus': "audio/ogg; codecs=opus; rate=24000";
+    readonly 'ogg-48khz-16bit-mono-opus': "audio/ogg; codecs=opus; rate=48000";
+    readonly 'webm-16khz-16bit-mono-opus': "audio/webm; codec=opus";
+    readonly 'webm-24khz-16bit-mono-opus': "audio/webm; codec=opus";
+    readonly 'webm-24khz-16bit-24kbps-mono-opus': "audio/webm; codec=opus";
+    readonly 'amr-wb-16000hz': "audio/amr-wb";
+    readonly 'g722-16khz-64kbps': "audio/g722";
+};
+export declare const AUDIO_FORMATS_NON_STREAM: {
+    readonly 'riff-8khz-8bit-mono-alaw': "audio/x-wav";
+    readonly 'riff-8khz-8bit-mono-mulaw': "audio/x-wav";
+    readonly 'riff-8khz-16bit-mono-pcm': "audio/x-wav";
+    readonly 'riff-22050hz-16bit-mono-pcm': "audio/x-wav";
+    readonly 'riff-24khz-16bit-mono-pcm': "audio/x-wav";
+    readonly 'riff-44100hz-16bit-mono-pcm': "audio/x-wav";
+    readonly 'riff-48khz-16bit-mono-pcm': "audio/x-wav";
+};
+type AudioOutputFormat = keyof typeof AUDIO_FORMATS_STREAM;
+declare const ALL_AUDIO_FORMATS: {
+    'riff-8khz-8bit-mono-alaw': "audio/x-wav";
+    'riff-8khz-8bit-mono-mulaw': "audio/x-wav";
+    'riff-8khz-16bit-mono-pcm': "audio/x-wav";
+    'riff-22050hz-16bit-mono-pcm': "audio/x-wav";
+    'riff-24khz-16bit-mono-pcm': "audio/x-wav";
+    'riff-44100hz-16bit-mono-pcm': "audio/x-wav";
+    'riff-48khz-16bit-mono-pcm': "audio/x-wav";
+    'raw-8khz-8bit-mono-alaw': "audio/basic";
+    'raw-8khz-8bit-mono-mulaw': "audio/basic";
+    'raw-8khz-16bit-mono-pcm': "audio/basic";
+    'raw-16khz-16bit-mono-pcm': "audio/basic";
+    'raw-22050hz-16bit-mono-pcm': "audio/basic";
+    'raw-24khz-16bit-mono-pcm': "audio/basic";
+    'raw-44100hz-16bit-mono-pcm': "audio/basic";
+    'raw-48khz-16bit-mono-pcm': "audio/basic";
+    'raw-16khz-16bit-mono-truesilk': "audio/SILK";
+    'raw-24khz-16bit-mono-truesilk': "audio/SILK";
+    'audio-16khz-32kbitrate-mono-mp3': "audio/mpeg";
+    'audio-16khz-64kbitrate-mono-mp3': "audio/mpeg";
+    'audio-16khz-128kbitrate-mono-mp3': "audio/mpeg";
+    'audio-24khz-48kbitrate-mono-mp3': "audio/mpeg";
+    'audio-24khz-96kbitrate-mono-mp3': "audio/mpeg";
+    'audio-24khz-160kbitrate-mono-mp3': "audio/mpeg";
+    'audio-48khz-96kbitrate-mono-mp3': "audio/mpeg";
+    'audio-48khz-192kbitrate-mono-mp3': "audio/mpeg";
+    'audio-16khz-16bit-32kbps-mono-opus': "audio/opus";
+    'audio-24khz-16bit-24kbps-mono-opus': "audio/opus";
+    'audio-24khz-16bit-48kbps-mono-opus': "audio/opus";
+    'ogg-16khz-16bit-mono-opus': "audio/ogg; codecs=opus; rate=16000";
+    'ogg-24khz-16bit-mono-opus': "audio/ogg; codecs=opus; rate=24000";
+    'ogg-48khz-16bit-mono-opus': "audio/ogg; codecs=opus; rate=48000";
+    'webm-16khz-16bit-mono-opus': "audio/webm; codec=opus";
+    'webm-24khz-16bit-mono-opus': "audio/webm; codec=opus";
+    'webm-24khz-16bit-24kbps-mono-opus': "audio/webm; codec=opus";
+    'amr-wb-16000hz': "audio/amr-wb";
+    'g722-16khz-64kbps': "audio/g722";
+};
+declare const FORMAT_CONTENT_TYPE: Map<string, "audio/basic" | "audio/SILK" | "audio/mpeg" | "audio/opus" | "audio/ogg; codecs=opus; rate=16000" | "audio/ogg; codecs=opus; rate=24000" | "audio/ogg; codecs=opus; rate=48000" | "audio/webm; codec=opus" | "audio/amr-wb" | "audio/g722" | "audio/x-wav">;
+export { ALL_AUDIO_FORMATS, FORMAT_CONTENT_TYPE, AudioOutputFormat };

package/dist/config/formats.js

@@ -0,0 +1,54 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.FORMAT_CONTENT_TYPE = exports.ALL_AUDIO_FORMATS = exports.AUDIO_FORMATS_NON_STREAM = exports.AUDIO_FORMATS_STREAM = void 0;
+exports.AUDIO_FORMATS_STREAM = {
+    // RAW PCM
+    'raw-8khz-8bit-mono-alaw': 'audio/basic',
+    'raw-8khz-8bit-mono-mulaw': 'audio/basic',
+    'raw-8khz-16bit-mono-pcm': 'audio/basic',
+    'raw-16khz-16bit-mono-pcm': 'audio/basic',
+    'raw-22050hz-16bit-mono-pcm': 'audio/basic',
+    'raw-24khz-16bit-mono-pcm': 'audio/basic',
+    'raw-44100hz-16bit-mono-pcm': 'audio/basic',
+    'raw-48khz-16bit-mono-pcm': 'audio/basic',
+    // TRUESILK
+    'raw-16khz-16bit-mono-truesilk': 'audio/SILK',
+    'raw-24khz-16bit-mono-truesilk': 'audio/SILK',
+    // MP3
+    'audio-16khz-32kbitrate-mono-mp3': 'audio/mpeg',
+    'audio-16khz-64kbitrate-mono-mp3': 'audio/mpeg',
+    'audio-16khz-128kbitrate-mono-mp3': 'audio/mpeg',
+    'audio-24khz-48kbitrate-mono-mp3': 'audio/mpeg',
+    'audio-24khz-96kbitrate-mono-mp3': 'audio/mpeg',
+    'audio-24khz-160kbitrate-mono-mp3': 'audio/mpeg',
+    'audio-48khz-96kbitrate-mono-mp3': 'audio/mpeg',
+    'audio-48khz-192kbitrate-mono-mp3': 'audio/mpeg',
+    // OPUS in container
+    'audio-16khz-16bit-32kbps-mono-opus': 'audio/opus',
+    'audio-24khz-16bit-24kbps-mono-opus': 'audio/opus',
+    'audio-24khz-16bit-48kbps-mono-opus': 'audio/opus',
+    // OGG
+    'ogg-16khz-16bit-mono-opus': 'audio/ogg; codecs=opus; rate=16000',
+    'ogg-24khz-16bit-mono-opus': 'audio/ogg; codecs=opus; rate=24000',
+    'ogg-48khz-16bit-mono-opus': 'audio/ogg; codecs=opus; rate=48000',
+    // WEBM
+    'webm-16khz-16bit-mono-opus': 'audio/webm; codec=opus',
+    'webm-24khz-16bit-mono-opus': 'audio/webm; codec=opus',
+    'webm-24khz-16bit-24kbps-mono-opus': 'audio/webm; codec=opus',
+    'amr-wb-16000hz': 'audio/amr-wb',
+    'g722-16khz-64kbps': 'audio/g722'
+};
+exports.AUDIO_FORMATS_NON_STREAM = {
+    // RIFF (WAV)
+    'riff-8khz-8bit-mono-alaw': 'audio/x-wav',
+    'riff-8khz-8bit-mono-mulaw': 'audio/x-wav',
+    'riff-8khz-16bit-mono-pcm': 'audio/x-wav',
+    'riff-22050hz-16bit-mono-pcm': 'audio/x-wav',
+    'riff-24khz-16bit-mono-pcm': 'audio/x-wav',
+    'riff-44100hz-16bit-mono-pcm': 'audio/x-wav',
+    'riff-48khz-16bit-mono-pcm': 'audio/x-wav',
+};
+const ALL_AUDIO_FORMATS = { ...exports.AUDIO_FORMATS_STREAM, ...exports.AUDIO_FORMATS_NON_STREAM };
+exports.ALL_AUDIO_FORMATS = ALL_AUDIO_FORMATS;
+const FORMAT_CONTENT_TYPE = new Map(Object.entries(ALL_AUDIO_FORMATS));
+exports.FORMAT_CONTENT_TYPE = FORMAT_CONTENT_TYPE;

package/dist/services/EdgeTTS.d.ts

@@ -16,6 +16,8 @@ export declare class EdgeTTS {
     private audio_format;
     private ws;
     getVoices(): Promise<Voice[]>;
+    getVoicesByLanguage(locale: string): Promise<Voice[]>;
+    getVoicesByGender(gender: 'Male' | 'Female'): Promise<Voice[]>;
     private generateUUID;
     private validatePitch;
     private validateRate;
@@ -23,8 +25,16 @@ export declare class EdgeTTS {
     synthesize(text: string, voice?: string, options?: SynthesisOptions): Promise<void>;
     private getSSML;
     private buildTTSConfigMessage;
+    synthesizeStream(text: string, voice?: string, options?: SynthesisOptions): AsyncGenerator<Uint8Array, void, unknown>;
     private processAudioData;
-    toFile(outputPath: string): Promise<string>;
+    private generateSecMsGec;
+    getDuration(): number;
+    getAudioInfo(): {
+        size: number;
+        format: string;
+        estimatedDuration: number;
+    };
+    toFile(outputPath: string, format?: string): Promise<string>;
     toRaw(): string;
     toBase64(): string;
     toBuffer(): Buffer;

package/dist/services/EdgeTTS.js

@@ -28,7 +28,11 @@ class EdgeTTS {
     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(`${constants_1.Constants.VOICES_URL}?trustedclienttoken=${constants_1.Constants.TRUSTED_CLIENT_TOKEN}`, {
+            headers: {
+                "User-Agent": constants_1.Constants.USER_AGENT
+            },
+        });
         const data = await response.json();
         return data.map((voice) => {
             delete voice.VoiceTag;
@@ -37,6 +41,14 @@ class EdgeTTS {
             return voice;
         });
     }
+    async getVoicesByLanguage(locale) {
+        const voices = await this.getVoices();
+        return voices.filter(voice => voice.Locale.startsWith(locale));
+    }
+    async getVoicesByGender(gender) {
+        const voices = await this.getVoices();
+        return voices.filter(voice => voice.Gender === gender);
+    }
     generateUUID() {
         return 'xxxxxxxx-xxxx-xxxx-yxxx-xxxxxxxxxxxx'.replace(/[xy]/g, function (c) {
             const r = Math.random() * 16 | 0;
@@ -84,10 +96,16 @@ class EdgeTTS {
         return `${volumeValue}%`;
     }
     async synthesize(text, voice = 'en-US-AnaNeural', options = {}) {
+        const secMsGEC = await this.generateSecMsGec(constants_1.Constants.TRUSTED_CLIENT_TOKEN);
         return new Promise((resolve, reject) => {
             this.audio_stream = [];
             const req_id = this.generateUUID();
-            this.ws = new ws_1.default(`${constants_1.Constants.WSS_URL}?trustedclienttoken=${constants_1.Constants.TRUSTED_CLIENT_TOKEN}&ConnectionId=${req_id}`);
+            const url = `${constants_1.Constants.WSS_URL}?TrustedClientToken=${constants_1.Constants.TRUSTED_CLIENT_TOKEN}&Sec-MS-GEC=${secMsGEC}&Sec-MS-GEC-Version=1-130.0.2849.68&ConnectionId=${req_id}`;
+            this.ws = new ws_1.default(url, {
+                headers: {
+                    "User-Agent": constants_1.Constants.USER_AGENT
+                }
+            });
             const SSML_text = this.getSSML(text, voice, options);
             const timeout = setTimeout(() => {
                 if (this.ws && this.ws.readyState === ws_1.default.OPEN) {
@@ -130,6 +148,79 @@ class EdgeTTS {
         return `X-Timestamp:${new Date().toISOString()}Z\r\nContent-Type:application/json; charset=utf-8\r\nPath:speech.config\r\n\r\n` +
             `{"context":{"synthesis":{"audio":{"metadataoptions":{"sentenceBoundaryEnabled":false,"wordBoundaryEnabled":true},"outputFormat":"audio-24khz-48kbitrate-mono-mp3"}}}}`;
     }
+    async *synthesizeStream(text, voice = 'en-US-AnaNeural', options = {}) {
+        this.audio_stream = [];
+        const req_id = this.generateUUID();
+        const secMsGEC = await this.generateSecMsGec(constants_1.Constants.TRUSTED_CLIENT_TOKEN);
+        const url = `${constants_1.Constants.WSS_URL}?TrustedClientToken=${constants_1.Constants.TRUSTED_CLIENT_TOKEN}&Sec-MS-GEC=${secMsGEC}&Sec-MS-GEC-Version=1-130.0.2849.68&ConnectionId=${req_id}`;
+        this.ws = new ws_1.default(url);
+        const SSML_text = this.getSSML(text, voice, options);
+        const queue = [];
+        let done = false;
+        let error = null;
+        let notify = null;
+        const push = (chunk) => {
+            queue.push(chunk);
+            if (notify) {
+                notify();
+                notify = null;
+            }
+        };
+        const timeout = setTimeout(() => {
+            if (this.ws && this.ws.readyState === ws_1.default.OPEN) {
+                this.ws.close();
+            }
+        }, 30000);
+        this.ws.on('open', () => {
+            const message = this.buildTTSConfigMessage();
+            this.ws.send(message);
+            const speechMessage = `X-RequestId:${req_id}\r\nContent-Type:application/ssml+xml\r\nX-Timestamp:${new Date().toISOString()}Z\r\nPath:ssml\r\n\r\n${SSML_text}`;
+            this.ws.send(speechMessage);
+        });
+        this.ws.on('message', (data) => {
+            const buffer = ensureBuffer(data);
+            const needle = buffer_1.Buffer.from('Path:audio\r\n');
+            const audioStartIndex = buffer.indexOf(new Uint8Array(needle));
+            if (audioStartIndex !== -1) {
+                const audioChunk = buffer.subarray(audioStartIndex + needle.length);
+                const chunk = new Uint8Array(audioChunk);
+                this.audio_stream.push(chunk);
+                push(chunk);
+            }
+            if (buffer.toString().includes('Path:turn.end')) {
+                this.ws?.close();
+            }
+        });
+        this.ws.on('error', (err) => {
+            error = err;
+            done = true;
+            if (notify) {
+                notify();
+                notify = null;
+            }
+        });
+        this.ws.on('close', () => {
+            clearTimeout(timeout);
+            done = true;
+            if (notify) {
+                notify();
+                notify = null;
+            }
+        });
+        while (!done || queue.length > 0) {
+            if (queue.length === 0) {
+                await new Promise(resolve => (notify = resolve));
+                continue;
+            }
+            const chunk = queue.shift();
+            if (chunk) {
+                yield chunk;
+            }
+        }
+        if (error) {
+            throw error;
+        }
+    }
     processAudioData(data) {
         const buffer = ensureBuffer(data);
         const needle = buffer_1.Buffer.from("Path:audio\r\n");
@@ -142,9 +233,40 @@ class EdgeTTS {
             this.ws?.close();
         }
     }
-    async toFile(outputPath) {
+    async generateSecMsGec(trustedClientToken) {
+        const ticks = Math.floor(Date.now() / 1000) + 11644473600;
+        const rounded = ticks - (ticks % 300);
+        const windowsTicks = rounded * 10000000;
+        const encoder = new TextEncoder();
+        const data = encoder.encode(`${windowsTicks}${trustedClientToken}`);
+        const hashBuffer = await crypto.subtle.digest('SHA-256', data);
+        return Array.from(new Uint8Array(hashBuffer))
+            .map(b => b.toString(16).padStart(2, '0'))
+            .join('')
+            .toUpperCase();
+    }
+    getDuration() {
+        if (this.audio_stream.length === 0) {
+            throw new Error("No audio data available");
+        }
+        // Estimate duration based on the size of the audio stream
+        const bufferSize = this.toBuffer().length;
+        const estimatedDuration = bufferSize / (24000 * 3); // 24000 Hz sample rate, 3 bytes per sample (16-bit stereo)
+        return estimatedDuration;
+    }
+    getAudioInfo() {
+        const buffer = this.toBuffer();
+        return {
+            size: buffer.length,
+            format: this.audio_format,
+            estimatedDuration: this.getDuration()
+        };
+    }
+    async toFile(outputPath, format = this.audio_format) {
+        if (!format || typeof format !== 'string')
+            format = this.audio_format;
         const audioBuffer = this.toBuffer();
-        const finalPath = `${outputPath}.${this.audio_format}`;
+        const finalPath = `${outputPath}.${format}`;
         await (0, promises_1.writeFile)(finalPath, new Uint8Array(audioBuffer));
         return finalPath;
     }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@andresaya/edge-tts",
-  "version": "1.3.0",
+  "version": "1.3.2",
   "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": [