⚙️ WasmTools

Text to Speech Online — Free Natural Voices, 100% In-Browser

Free online text to speech: 174 natural neural voices across 55 languages, powered by Piper TTS running 100% in your browser. No upload, no signup — download the audio as WAV.

🔒 Runs entirely in your browser — your files never leave your device.

About Text to Speech

Convert text to speech with natural neural voices, entirely on your device. This tool runs Piper — a fast, open-source speech synthesis system — inside your browser as WebAssembly: your text is phonemized locally, fed to a voice model, and returned as audio you can play or download as WAV. Voice models (20–130 MB each) are downloaded once from the open Piper catalog on Hugging Face and cached by your browser; the text you type is never transmitted anywhere.

How to use Text to Speech

  1. Pick a language and a voice — the size shown is what will be downloaded on first use.
  2. Type or paste your text and adjust the speed if you like.
  3. Press Speak — the first run downloads the voice model once; after that it starts instantly.
  4. Listen to the result and download it as a WAV file.

Neural text to speech, entirely in your browser

How it works

Piper is a fast neural speech synthesis system based on the VITS architecture. This page runs the whole pipeline locally: your text is first converted to phonemes by espeak-ng (compiled to WebAssembly), the phonemes are fed to the voice's ONNX model through the onnxruntime WebAssembly engine, and the result is 16-bit WAV audio.

There is no server round trip. The voice model is the only download, and your browser caches it after the first use.

Choosing a voice

  • Basic (x_low) voices are ~20 MB and download fastest; Medium and High voices sound clearer and more natural.
  • Every voice you use stays cached, so trying several costs nothing after the first download.

Speed and quality tips

The speed slider changes speaking rate without affecting pitch. Values above 1.0× speak faster.

Short paragraphs synthesize best; very long texts can exhaust browser memory, so split them.

Numbers, abbreviations and URLs are read more reliably when written out as words.

Why local TTS matters

Reading your own documents aloud — drafts, emails, study notes — shouldn't require handing the text to a cloud API. Local synthesis keeps private content private, works offline after the first download, and has no usage limits or accounts.

This tool pairs with the Audio Transcriber on this site: one turns speech into text, this one turns text back into speech — both entirely on your device.

Frequently asked questions

Is my text uploaded anywhere?

No. The voice model runs locally in your browser as WebAssembly. The only thing downloaded is the voice model itself (once) from the open Piper catalog — your text never leaves your device.

Why is the first run slow?

Each voice model is 20–130 MB and is downloaded the first time you use it. Your browser caches it afterwards, so later runs start instantly. Smaller "Basic" voices download fastest.

Which languages are supported?

55 locales with 174 voices in total — including English (US/UK/AU), Chinese, Spanish, French, German, Russian, Japanese, Arabic, Hindi, Portuguese and many more.

How much disk space does it use?

Each voice you use is cached separately (20–130 MB). You can switch between voices freely; to free space, clear the site data for this page in your browser settings.

Can I use the audio commercially?

The Piper voices are MIT-licensed and the audio you generate is yours. Check the individual voice's model card for any attribution norms.

Which browsers work?

Any modern Chrome, Edge, Firefox or Safari. Synthesis is single-threaded, so no special browser settings or headers are needed.

Related tools

Audio Converter

Convert audio files to WAV or MP3 entirely in your browser.

runs locally

Audio Cutter

Trim an audio file to a start and end time.

runs locally

Audio Noise Reducer

Reduce background hiss and noise from recordings with a noise gate.

runs locally

Voice Recorder

Record audio from your microphone and download it as a WAV file.

runs locally

Audio Speed Changer

Speed up or slow down audio without changing its pitch.

runs locally

Audio Transcriber

Turn voice recordings, interviews, and voicemails into text with a local Whisper model — no audio ever leaves your device.

runs locally