Text to Speech

217 / 20,000 characters

More info

Long text is read sentence by sentence because some browsers stop a single utterance after about 15 seconds. Changes to the voice, speed, pitch or volume apply from the next part. Word highlighting depends on the voice; some voices, including many online ones, do not report word positions.

Built-in voices run on your device. Voices marked online are provided by your browser or operating system vendor, which may send the text to its servers to generate the speech.

About this tool

This text-to-speech tool reads any text aloud with the voices built into your browser and operating system. Paste an article, an email draft or study notes, pick a voice in your language, and adjust the speed, pitch and volume. The word being spoken is highlighted so you can follow along, which helps with proofreading, language learning and resting your eyes during long reading sessions.

The voices come from your device, so the selection differs between Windows, macOS, iOS, Android and Linux, and between browsers. Voices are grouped by language, and the list starts with voices for the language of this page. The tool works without an account and does not save your text.

How it works

Type or paste up to 20,000 characters, choose a voice and press Play. Use Pause and Resume to take a break, or Stop to end playback and edit the text. Speed ranges from 0.5× to 2×, pitch from 0 to 2 and volume from 0 to 100%.

The tool uses the Web Speech API (speechSynthesis) built into the browser. It splits the text into sentences and groups them into parts of about 200 characters at normal speed (fewer when slower, more when faster), because some speech engines cut off a single utterance after roughly 15 seconds. While a part is spoken, the browser reports the position of each word, and the tool highlights it in the reading view.

Example

A 1,000-word article is roughly 6,000 characters. At normal speed most voices read about 150 to 180 words per minute, so the article takes around 6 minutes, or about 4 minutes at 1.5×. The tool reads it in about 30 parts, one after another, without pauses you would normally notice.

To proofread an email, paste it, choose a voice in the language of the email and set the speed to 0.9×. Listening while following the highlight makes missing words, repeated words and awkward sentences easier to catch than reading silently.

Assumptions and limits

Speech is generated by your browser and operating system, so voices, pronunciation and word highlighting vary between devices. Some voices, often labeled online, are generated on the vendor's servers and need an internet connection. The Web Speech API cannot record or export audio, so this tool has no MP3 download. Mobile browsers may stop speech when the screen locks or the tab goes to the background. Text is limited to 20,000 characters and is not saved.

Frequently asked questions

Can I download the speech as an MP3?

No. Browsers play synthesized speech directly through your speakers and do not give web pages access to the audio, so it cannot be saved as MP3 or WAV here. For an audio file, use a text-to-speech app that offers audio export, or record the playback with an audio recorder.

Why are the voices different on my phone and my computer?

The voices belong to the operating system and the browser, not to this website. Windows, macOS, iOS, Android and ChromeOS each ship their own voices, and Chrome and Edge add online voices. You can usually install more voices in the system's language or accessibility settings.

Is my text sent anywhere?

Not by this tool: it does not upload or store your text. Voices that run on your device process the text locally. Voices marked online, such as the Google voices in Chrome or the Microsoft online voices in Edge, are generated on the vendor's servers, so the browser sends the text there. Choose a local voice for confidential text.

Why does reading stop or skip?

Mobile browsers often pause speech when the screen turns off or you switch apps. Some engines also stop long utterances, which is why this tool reads in short parts. If a voice keeps failing, choose another one or reload the page.

Why is no word highlighted?

Highlighting relies on word boundary events from the speech engine. Many online voices and some Android voices do not send them, so the speech plays normally but the highlight stays at the start of each part.

Link with your values

Add these parameters to this page’s address to open it with the values filled in, for example to share a calculation or to link to it from an AI assistant. The result is calculated in the browser: the page HTML shows an example for the default values, and the tool sets the data-url-params attribute once the link’s values are applied.

ParameterValueIf omitted
textText to readtext; your own text; Copy link includes itrequired
voiceVoicetext; your own text; Copy link includes itoptional
rateSpeednumber; allowed 0.5–2default 1
pitchPitchnumber; allowed 0–2default 1
volume_pctVolumepercent: 6.5 means 6.5%; allowed 0–100default 100

Examples

  • ?text=Hello%2C%20world.%20This%20page%20reads%20text%20aloud.&rate=1.2
  • ?text=Good%20morning&pitch=1.5&volume_pct=60

Write decimals with a point. Encode a space as %20, + as %2B, # as %23 and & as %26.

Result elements: [data-result-block="speech"]