Edge TTS Server-Side Package

Server-side implementation of the Edge TTS package — utilizing the Microsoft Edge Read Aloud API to generate text-to-speech with support for data streaming or Base64 encoding.

Features

• Text-to-Speech Conversion: Effortlessly convert text into natural-sounding speech using Microsoft Edge's TTS capabilities.
• Voice Customization: Access a wide range of voices and adjust rate, volume, and pitch to tailor the output for your project.
• Flexible Audio Output: Export synthesized audio in various formats including streaming, raw audio, or Base64-encoded output.

Installation

Install the Edge TTS package via npm or bun by running one of the following commands:

npm install @devsobrinho/edge-tts

yarn add @devsobrinho/edge-tts

bun add @devsobrinho/edge-tts

Configuration Options

The EdgeTTS package allows for configuration either globally or on a per-method basis:

• Global Configuration: Pass a configuration object to the constructor for a persistent setup.
• Dynamic Method Configuration: Optionally override global settings by passing a configuration object to individual methods (toRaw, toBase64, or toStream).

The EdgeTTSSynthesizeConfig interface defines the following configurable options:

export interface EdgeTTSSynthesizeConfig {
  voice?: string;
  rate: RATE | string | number;
  volume: VOLUME | string | number;
  pitch: PITCH | string;
  voiceLocale?: string;
  outputFormat?: OUTPUT_FORMAT;
}

Usage Examples

Streaming Output

The following example shows how to stream the generated audio data:

import { EdgeTTS } from "@devsobrinho/edge-tts";
import { PassThrough } from "stream";

const abortController = new AbortController();
const stream = new PassThrough({ signal: abortController.signal });
const edgeTTS = new EdgeTTS();

await edgeTTS.toStream(
  stream,
  "Hello, my name is Francisca, and I am your virtual assistant.",
  {
    voice:
      "Microsoft Server Speech Text to Speech Voice (pt-BR, FranciscaNeural)",
    volume: "50%",
  }
);

stream.on("data", (data) => {
  console.log("Data chunk received:", data);
});

stream.on("close", () => {
  console.log("Stream closed.");
});

stream.on("error", (err) => {
  console.error("Stream error:", err);
});

Base64 Output

For projects where Base64 encoding is more suitable, here’s an example:

import { EdgeTTS } from "@devsobrinho/edge-tts";

const edgeTTS = new EdgeTTS();

const base64Audio = await edgeTTS.toBase64(
  "Hello, my name is Francisca, and I am your virtual assistant.",
  {
    voice:
      "Microsoft Server Speech Text to Speech Voice (pt-BR, FranciscaNeural)",
    volume: "50%",
  }
);

console.log("Base64 Audio:", base64Audio);

Acknowledgments

We would like to thank the developers and contributors of related projects for their valuable groundwork and inspiration, including:

• Examples from rany2/edge-tts
• Utility functions in rany2/edge-tts
• HASS Edge TTS Integration