@ziplayer/tts

A TypeScript package for text-to-speech with two providers:

| Provider | Returns | Notes | | ------------------------ | ----------------------------- | ------------------------------------- | | Google Translate TTS | string[] (audio URLs) | No auth required; max 200 chars/chunk | | Microsoft Edge TTS | Promise<Buffer> (raw audio) | 400+ neural voices; WebSocket-based |

Installation

npm install @ziplayer/tts

Requires Node.js ≥ 18 (uses native fetch + ws).

Google TTS — getTTSUrls

Returns an array of Google Translate audio URLs. Long texts are automatically split into ≤ 200-character chunks.

import { getTTSUrls } from "@ziplayer/tts";

const urls = getTTSUrls("Hello! This is a test.", { lang: "en" });
console.log(urls);
// ["https://translate.google.com/translate_tts?..."]

Options — GoogleTTSOptions

| Property | Type | Default | Description | | -------- | --------- | ------- | ------------------------------------------------ | | lang | string | "en" | BCP-47 language code ("vi", "ja", "ko", …) | | slow | boolean | false | Speak at a slower pace |

Edge TTS — getEdgeTTS

Synthesises text via the Microsoft Edge Read Aloud API and returns a raw audio Buffer. Supports 400+ neural voices across many languages.

import { getEdgeTTS } from "@ziplayer/tts";
import { writeFileSync } from "fs";

const audio = await getEdgeTTS("こんにちは！", {
	voice: "ja-JP-NanamiNeural",
	rate: "+10%",
	pitch: "+0Hz",
});

writeFileSync("output.mp3", audio);

Options — EdgeTTSOptions

| Property | Type | Default | Description | | -------------- | ------------------ | ----------------------------------- | ---------------------------------------- | | voice | string | "en-US-AriaNeural" | Voice short name (see voices list below) | | rate | string | "+0%" | Speaking rate, e.g. "+20%", "-10%" | | pitch | string | "+0Hz" | Pitch, e.g. "+5Hz", "-10Hz" | | volume | string | "+0%" | Volume, e.g. "+10%", "-20%" | | outputFormat | EdgeOutputFormat | "audio-24khz-48kbitrate-mono-mp3" | See formats below |

Output Formats

audio-16khz-32kbitrate-mono-mp3   audio-16khz-64kbitrate-mono-mp3
audio-16khz-128kbitrate-mono-mp3  audio-24khz-48kbitrate-mono-mp3  ← default
audio-24khz-96kbitrate-mono-mp3   audio-24khz-160kbitrate-mono-mp3
audio-48khz-96kbitrate-mono-mp3   audio-48khz-192kbitrate-mono-mp3
webm-24khz-16bit-mono-opus        ogg-24khz-16bit-mono-opus
raw-24khz-16bit-mono-pcm          ...

Edge TTS — getEdgeTTSVoices

Fetches the full list of available voices from Microsoft.

import { getEdgeTTSVoices } from "@ziplayer/tts";

const voices = await getEdgeTTSVoices();

// Filter by locale
const japanese = voices.filter((v) => v.Locale.startsWith("ja-"));
const female = voices.filter((v) => v.Gender === "Female");

console.log(voices[0]);
// {
//   Name: "Microsoft Server Speech Text to Speech Voice (en-US, AriaNeural)",
//   ShortName: "en-US-AriaNeural",
//   Gender: "Female",
//   Locale: "en-US",
//   FriendlyName: "Microsoft Aria Online (Natural) - English (United States)",
//   ...
// }

Popular Voices

| Language | Voice | Gender | | ------------------ | ---------------------- | ------ | | English (US) | en-US-AriaNeural | Female | | English (US) | en-US-GuyNeural | Male | | English (GB) | en-GB-SoniaNeural | Female | | Vietnamese | vi-VN-HoaiMyNeural | Female | | Japanese | ja-JP-NanamiNeural | Female | | Korean | ko-KR-SunHiNeural | Female | | Chinese (Mandarin) | zh-CN-XiaoxiaoNeural | Female |

TypeScript Types

All types are exported:

import type { GoogleTTSOptions, EdgeTTSOptions, EdgeTTSVoice, EdgeOutputFormat } from "@ziplayer/tts";

License

MIT © Ziji