node-speaknow

A simple Node.js package for text-to-speech (TTS) using system voices (Windows/macOS). Supports sequential speech so multiple messages don’t overlap.

Installation

npm install node-speaknow

Usage

JavaScript Example

const NodeSpeak = require("node-speaknow");

(async () => {
  const tts = new NodeSpeak();

  // Get available voices
  const voices = await tts.getVoices();
  console.log("Available voices:", voices);

  // Set a voice (optional, defaults to system default)
  await tts.setVoice("Microsoft Zira Desktop");

  // Speak messages sequentially
  tts.speakNow("Hello world!");
  tts.speakNow("This will play after the first message.");
  tts.speakNow("And this one comes third.");
})();

TypeScript Example

import NodeSpeak from "node-speak";

(async () => {
  const tts = new NodeSpeak();

  await tts.setVoice("Microsoft Zira Desktop");

  tts.speakNow("Hello from TypeScript!");
})();

🖥️ Supported Platforms

• Windows → Uses System.Speech.Synthesis.SpeechSynthesizer via PowerShell
• macOS → Uses the built-in say command
• Linux → Not yet supported (planned)

⚡ Features

• Get available system voices
• Choose a specific voice
• Queue system → prevents overlapping speech
• Works in both JavaScript and TypeScript projects

📖 API Reference

new NodeSpeak(options?: { voice?: string })

Creates a new instance of the TTS engine.

| Option | Type | Default | Description | | ------ | ------ | -------------- | ------------------ | | voice | string | system default | Initial voice name |

getVoices(showLog?: boolean): Promise<string[]>

Returns a list of available voices.

setVoice(voiceName: string): Promise<void>

Sets the active voice. Must match one of the names from getVoices().

speakNow(text: string): void

Speaks the given text. If multiple calls are made, they are queued and played one after another.

License

MIT © 2025 Zainul Khan