Text to Speech
Paste any text and hear it read aloud with your device’s voices. No character limit, no signup, speed and pitch control, and sentence highlighting.
245 characters · 47 words · about 19 s at 1×
What Text to Speech Does
This text to speech tool reads anything you paste in a real voice, using the speech engine that already ships with your browser and operating system. There is no account and no character quota, because the speech comes from the voices your browser and device already provide, not from a paid service run by this site. Paste a single paragraph or an entire chapter and it reads straight through.
Voices are the part that varies. A Windows laptop, a Mac, an Android phone and an iPhone each offer their own set, and Chrome and Edge add extra online voices on top. Each voice in the list is labeled either as installed on your device or as “online”. That label matters for privacy: the Web Speech standard lets a browser use a remote synthesizer, and when it does, the text you paste travels to that provider to be turned into audio.
Long passages are read one sentence-sized chunk at a time. That fixes the most common complaint about browser speech, Chrome’s online voices falling silent after a few sentences, and it lets the page highlight exactly which sentence is being spoken, which makes following along or proofreading by ear much easier.
How to Use Text to Speech
- Type or paste the text you want to hear into the box
- Choose a language and a voice. Voices marked “online” send text to a speech provider
- Set the speed and pitch with the sliders
- Press Play; the sentence being read is highlighted as it goes
- Use Pause, Resume or Stop at any time
How Text to Speech Works
Speaking long text reliably
collapse whitespace → split into sentences at . ! ? ; : → pack sentences into chunks of at most about 220 characters (a longer sentence is cut at a comma, else a space) → speak chunk by chunk with SpeechSynthesisUtterance at the chosen voice, rate (0.5–2) and pitch (0–2)
- chunks
- Some browsers stop a single long utterance partway through without an error. Short chunks avoid that and let the page highlight the part being read.
- voices
- Supplied by your operating system and browser, so the list differs between Windows, macOS, Android and iOS. Voices marked online, such as Chrome's Google voices, send the text to the browser vendor to be spoken; the others run on your device.
Worked example
A 499-character paragraph of four sentences, one of them about 280 characters long
- The first two sentences and the clause ending in a colon fit together: chunk 1, 182 characters
- The long sentence is over 220, so it is cut at the last comma before 220: chunk 2, 190 characters
- Its remainder and the short final sentence form chunk 3, 125 characters
Result: Three utterances spoken back to back, each short enough that the browser finishes it, with the current one highlighted.
Where the Voices Come From
The list you see is assembled by your browser from these sources. Adding voices at the operating-system level makes them appear here after a page reload.
| Platform | Built-in voices | Add more voices |
|---|---|---|
| Windows 10/11 | Microsoft desktop voices (e.g. David, Zira) | Settings › Time & language › Speech › Add voices |
| macOS | Apple system voices | System Settings › Accessibility › Spoken Content › System voice › Manage Voices |
| iPhone / iPad | Apple voices, including enhanced and premium downloads | Settings › Accessibility › Spoken Content › Voices |
| Android | Google speech services voices | Settings › Accessibility › Text-to-speech output |
| Chrome (any OS) | Adds several “Google …” online voices | Provided by the browser; marked online |
| Microsoft Edge | Adds Microsoft “Natural” online voices | Provided by the browser; marked online |
Reading Time at Different Speeds
Approximate listening time, assuming about 150 words per minute at 1×, a typical conversational speaking rate. Actual pace varies by voice.
| Length | At 1× | At 1.25× | At 1.5× |
|---|---|---|---|
| 300 words (a short email) | 2 min | 1 min 36 s | 1 min 20 s |
| 1,000 words (a blog post) | 6 min 40 s | 5 min 20 s | 4 min 27 s |
| 2,500 words (a long article) | 16 min 40 s | 13 min 20 s | 11 min 7 s |
| 5,000 words (a short story) | 33 min 20 s | 26 min 40 s | 22 min 13 s |
How to Read Your Result
Listening is a proofreading tool
Your eyes autocorrect your own writing; your ears do not. Hearing a draft read aloud at 1× exposes doubled words, missing articles and sentences that run out of breath far more reliably than rereading. It is the main reason writers use text to speech, and it works best with a local voice at a slightly slower speed.
Choosing between a local and an online voice
Online voices in Chrome and Edge usually sound more natural because they run on the provider’s servers. Local voices keep your text on your device and keep working offline. For confidential documents — contracts, medical notes, unpublished work — pick a voice without the “online” label.
Pitch and speed are independent
Speed changes how fast the words come; pitch raises or lowers the voice without changing its pace. Small pitch changes (0.9–1.1) can make a voice easier to listen to for long periods; large ones quickly sound artificial.
Limitations & Accuracy Notes
- Audio cannot be downloaded. Browser speech engines play straight to your speakers and do not hand web pages a copy of the sound, so no browser-based tool that uses them can produce an MP3.
- Voice quality and language coverage depend entirely on your device and browser; this site cannot add voices to the list.
- Pronunciation of names, acronyms and numbers follows the voice’s own rules. Spelling a word phonetically in the box is the only way to change how it is said.
- SSML markup (pauses, emphasis tags) is not supported; punctuation is the only way to shape pauses.
- Online voices need an internet connection and send the text to their provider; local voices work offline.
Frequently Asked Questions
Is this text to speech tool really free with no character limit?
Yes. It uses the speech engine that is already built into your browser and operating system, so there is no account, no credit system and no character cap. Very long text is spoken in sentence-sized pieces, one after another, so a whole article or chapter reads through without stopping.
Can I download the audio as an MP3?
No. The browser speech engine plays sound straight to your speakers and does not give web pages a copy of the audio, so no in-browser tool built on it can save a file. On a Mac, the built-in Terminal command say -o speech.aiff -f yourfile.txt exports spoken text; on other systems you need a dedicated text-to-speech app that writes files.
Why do I see different voices on my phone and my laptop?
The voices come from your operating system and browser, not from this website. Windows, macOS, iOS and Android each ship their own voices, Chrome adds several Google online voices, and Edge adds Microsoft’s online natural voices. You can install extra voices in your system’s speech or accessibility settings and they will appear in the list after a reload.
Is the text I paste sent to a server?
Not when you use a voice installed on your device, since the speech is generated locally. Voices marked “online” in the list are different: your browser sends the text to that provider (such as Google or Microsoft) to synthesize it. The tool tells you which kind the selected voice is so you can choose.
Why does speech stop partway through long text in some browsers?
Chrome’s online voices have a long-standing habit of going silent after roughly 15 seconds of continuous speech. This tool avoids it by speaking one sentence-sized chunk at a time and starting the next as soon as the previous one ends, which is also how it highlights the sentence being read.
How do I get more natural-sounding voices?
On Windows, open Settings › Time & language › Speech and add voices; on a Mac, go to System Settings › Accessibility › Spoken Content and manage system voices; on iPhone, Settings › Accessibility › Spoken Content › Voices. In Microsoft Edge, the voices labeled Natural are among the most realistic available to a web page.
What reading speed should I use?
For proofreading your own writing, 0.9× to 1× makes missing words and clumsy phrasing easy to hear. For listening to familiar material, 1.25× to 1.5× is comfortable for most people. Pitch changes the voice’s tone without changing its speed.
Does it read languages other than English?
Yes, in any language your device has a voice for. Use the Language filter to see which voices are available, then pick one that matches the text; an English voice reading Spanish text will mispronounce most words.
References & Further Reading
- MDN Web Docs — SpeechSynthesis — The browser API this tool uses
- MDN Web Docs — SpeechSynthesisVoice.localService — How a page tells local voices from remote ones