Lite Tools

Text to Speech

Read any text aloud with adjustable voice, rate and pitch.

Characters0

Voices come from your operating system and browser; the list may take a few seconds to load.

Speech is synthesized locally by your browser's built-in engine: the text never leaves your device. The number and quality of voices depend on the speech packages installed on your system.

How to use

  1. Type or paste your text into the box: the character count updates as you go.
  2. Choose a voice (grouped by language) and adjust the rate (0.5–2x), pitch (0–2) and volume (0–100%) sliders.
  3. Press "Play" to start; while reading you can "Pause", then "Resume" or "Stop" at any time.
  4. A status notice shows "Reading aloud…" and similar states, with feedback when reading ends or fails.
  5. Pressing "Play" again restarts from the beginning with your latest text and settings.

FAQ

Why is the voice list empty or very short?
Voices are provided by your OS and browser: Windows and macOS ship multi-language speech packs, while Android and iOS depend on system settings. The list loads asynchronously: wait a few seconds or reload the page. Installing more system language packs adds more voices.
Can I export the speech as an audio file?
The browser speech-synthesis API does not expose an audio stream, so this tool can only play it live. For an audio file, use your system's screen/audio recorder or a dedicated TTS service.
Is my text uploaded anywhere?
No. Synthesis happens locally in your browser's built-in engine. Neither the text nor your settings are sent to any server, so it is safe for drafts, emails and other private content.
Does reading continue in another tab?
Most desktop browsers keep the audio playing, but some mobile browsers pause or stop speech when the page goes to the background. Return to this page and press "Resume" or "Play" to continue.

How to use

  1. Enter the text to read

    Paste an article, email draft or vocabulary list into the text box. The character counter updates live so you can keep each reading at a sensible length.

  2. Pick a voice

    Choose a system voice from the Voice dropdown, grouped by language. The default follows the page language, and the list loads asynchronously, so give it a few seconds.

  3. Tune rate, pitch and volume

    Drag the Rate, Pitch and Volume sliders. Rate runs 0.5 to 2 times, pitch runs 0 to 2, and volume runs 0 to 100%. Fine-tune by ear until it sounds natural.

  4. Control playback while reading

    Press Play to start, then Pause, Resume or Stop at any point mid-read. Pressing Play again restarts from the beginning with your latest text and settings.

  5. Fix silence or failures

    If speech fails or stays silent, switch to another voice or lower the rate and retry. An empty voice list usually means the language pack is missing from your OS.

About this tool

This text-to-speech tool drives the Web Speech Synthesis engine built into your browser. It reads typed or pasted text aloud with a chosen voice at adjustable rate, pitch and volume. Synthesis happens entirely on your device, so drafts, emails and other private content never leave the page. There is no account, no upload and no waiting queue.

Writers use it to catch awkward sentences by ear. Language learners shadow pronunciation word by word, and busy readers get through long articles hands-free. Voice quality depends on the speech packages installed in your operating system. Windows and macOS ship multi-language voices out of the box, and installing further system language packs extends the choice.

How it works

The Web Speech API's synthesis interface is only a scheduler. The voices themselves come from the operating system: SAPI engines on Windows, AVSpeechSynthesizer on macOS and iOS, and speech-dispatcher on Linux desktops. That is why quality varies so widely. Decades-old formant synthesisers and recent neural voices can sit side by side in the same dropdown. The list also arrives asynchronously through a voiceschanged event, which is why it can populate a few seconds after the page opens. To compare two voices, play the same sentence with each one.

Rate 1 is the engine's natural speed. The underlying parameter actually spans 0.1 to 10, and this tool narrows it to a usable 0.5 to 2. Pitch 1 sits at the voice's natural centre. Browsers dropped SSML support years ago, so punctuation is the only prosody control left. Commas and periods are what make the engine pause, and an unpunctuated sentence tends to rush through. When the pacing sounds wrong, reset the rate to its default first and check whether the text or the setting is at fault.

When to use it

  1. Proofread by ear

    Silent reading lets the eye skip over errors; the ear does not. Play a finished email or draft straight through, and awkward phrasing, dropped words and duplicated phrases stand out at once.

  2. Shadowing for language learning

    Slow the rate and repeat each sentence after the voice, pausing to fix individual words. Imitating real prosody builds rhythm that phonetic charts alone never teach, and it fits a commute.

  3. Hands-free reading time

    During a commute, chores or a screen break, let a long article read itself. Install the language pack once, pick a pleasant voice, and listen for half an hour without tiring your eyes.

Tips & tricks

  • The browser API exposes no audio stream, so exporting a file needs a system screen/audio recorder or a dedicated TTS service.
  • Drop the rate to about 0.8x for English shadowing practice. A rate of 1.2x works well for quickly proofreading Chinese drafts.
  • Some mobile browsers pause speech in background tabs. Return to the page and press Resume to continue listening.

Things to watch

  • No audio file to export

    Speech is produced inside the OS engine, never routed through the page's audio graph. Capturing a file needs a system recorder, and there is no one-click export.

  • The voice list loads late

    Voices arrive asynchronously through the voiceschanged event, so the dropdown can be empty right after the page opens. If it stays empty, the OS is missing that language's speech pack.

  • Background tabs stop speech

    Many mobile browsers suspend synthesis the moment the tab backgrounds or the screen locks, as part of power throttling. Keep the page foregrounded for long listens and press Resume after interruptions.

Related Tools