Text to speech with the voices you already have
Press Speak and your browser reads the text aloud. This text to speech page uses the Web Speech API (speechSynthesis), so every voice in the list comes from the browser or from the operating system. Windows, macOS, Android and iOS ship different voices, and the list on a phone rarely matches the list on a laptop.
Proofreading is a good first job for it. A missing word is easier to hear than to see.
Speak, pause, resume, stop: the controls in order
- Type or paste your text into the box. The counter under the heading shows the length against the limit of 100,000 characters.
- Choose a voice. Voices for the language of the page come first, and inside a language the voices marked "on device" come before the "online" ones.
- Move the sliders for speed (0.5 to 2), pitch (0 to 2) and volume (0 to 1).
- Press Speak. Use Pause and Resume while it runs, or Stop to drop the rest.
Why a long text is read in parts
Some browsers cut an utterance off after about 15 seconds or a few hundred characters. So the tool splits your text at sentence ends (full stop, exclamation mark, question mark, an ellipsis, or the Chinese and Japanese 。) into parts of at most 200 characters and speaks them one after the other. A sentence above 200 characters is cut at spaces. A single word above 200 characters is cut inside the word.
The line under the buttons shows where you are, for example "Part 3 of 12". Pause and Resume work across the parts. Stop cancels all parts that are left. If you change voice, speed, pitch or volume while it speaks, the new value starts with the next part.
Which voices may send your text to a server?
The browser tells the page whether a voice is local. The tool shows this as "on device" or "online", and it selects an on-device voice by default. An online voice can make the browser or the operating system send the text to a server that creates the audio. The page cannot see this and cannot stop it. So choose an on-device voice for a draft contract or a private letter.
The tool itself sends nothing. Your text is not uploaded by this page, not stored, and not written to the clipboard. Closing or leaving the page stops the speech.
No audio file, and other hard limits
The Web Speech API plays sound but gives no access to it, so this tool cannot record or save an MP3 or WAV file. There is no download button for that reason. Text above 100,000 characters is refused with a message. A browser without speechSynthesis gets a message and disabled controls.
Some browsers load their voices late. The tool fills the list at once, fills it again when the browser reports new voices, and keeps your choice. If the list stays empty, the system has no voice installed.