How this text to speech tool works
The tool uses the Web Speech API, a speech interface that browsers offer to web pages, described in the Web Speech API specification, a W3C Community Group draft (this page follows the edition of 18 September 2026). The page passes your text to the browser's speech synthesis, and the browser reads it aloud with one of its voices. There is no account, no upload and no limit on the length of your text.
The text is read one sentence at a time. Line breaks start a new sentence too, so headings and list items are read on their own, and very long sentences are read in parts at a comma or a space. That makes it quick to jump around: click any sentence in the Read along view to start there, or use the arrow buttons to move to the previous or next sentence. Pause stops at the current word, and Resume continues from that word. To join lines that belong to one sentence before you listen, use remove line breaks.
While the voice speaks, the current sentence is shaded and the current word is marked. Word marking relies on position reports from the voice. The specification asks browsers to send these reports only when the speech engine provides them, so with some voices only the sentence is shaded. If you see that, the tool tells you below the buttons, and another voice may give you the word by word view.
Choosing a voice
The voice list comes from your browser and your operating system, sorted by language with your browser language first. The tool picks a voice that matches your browser language, and you can choose any other. So the number of voices, and how natural they sound, depends on your device and browser. The hint under the list tells you how many voices and languages yours offers.
A voice can come from a speech engine on your device or from a remote speech service. Browsers report this for every voice through a property named localService, and voices from a remote service carry the label online in the list. The specification notes that remote voices may bring extra delay, data use or cost. Because such a voice is synthesized outside your device, choose a voice without the online label for anything private.
Pick a voice in the language of your text. A voice reads every word with the pronunciation rules of its own language, so an English voice reading Spanish text sounds wrong even if every word is there. If your text mixes languages, read each part with a matching voice.
Speed, pitch and volume
The three sliders map directly onto the settings of the Web Speech API:
| Setting | On this page | What it does |
|---|---|---|
| Speed | 0.5× to 2× | How fast the voice reads, relative to its normal speed of 1. |
| Pitch | 0 to 2 | How high or low the voice sounds. 1 is the normal pitch of the voice. |
| Volume | 0% to 100% | How loud the voice is, on top of your device volume. |
The specification defines speed relative to the normal speed of each voice, so the same number can sound a little faster on one voice than on another. Changes apply right away: the voice starts again from the current word with the new setting. Tick Remember my voice and settings on this device to keep your choice for the next visit. It is saved only in your own browser, and unticking the box deletes it. Your text is never saved.
Why there is no MP3 download
In the Web Speech API, the browser plays the voice itself. The speechSynthesis interface has methods to speak, pause, resume and cancel and to list the voices, but none of them gives the page the audio. A website that uses these browser voices therefore cannot save the speech as an MP3 or WAV file, and this tool does not pretend otherwise: it plays the sound and nothing more. If you need a recording, you have to capture the sound from your device while it plays.
Ways to use it
- Proofread by ear: listen to an email, essay or post before you send it, and follow the highlight to find the spot you want to change. The readability checker highlights long sentences.
- Rest your eyes: listen to a long article or your notes instead of reading them on a screen.
- Practice a language: paste a sentence, pick a voice of that language and slow it down to hear each word.
- Time a script: hear how a voiceover or a speech sounds before you record it. The script timer works out how long it will take at your pace, and the word counter counts the words.
How to use the text to speech tool
- 1Type or paste your text into the box, or press Fill in an example to try it first.
- 2Pick a voice in the language of your text, and set the speed, pitch and volume.
- 3Press Play to listen. Follow the highlighted words, pause and resume, or click a sentence to jump there.
Frequently asked questions
Is this text to speech tool free?
Yes. It is free, with no sign up and no word limit. It uses the speech voices that your browser and operating system already provide, so there is nothing to install and nothing to pay for.
Can I download the speech as an MP3 file?
No. The Web Speech API that browsers use for this plays the voice directly through your speakers. Its speechSynthesis interface has methods to speak, pause, resume and cancel, but none that hands the audio to a website, so a page that uses these voices cannot save the speech as a file.
Why are the voices different on my phone and my computer?
The list comes from the browser you use and the device it runs on, so a laptop and a phone usually offer different voices, even on the same page. Pick the one you like best on each device.
What does online next to a voice mean?
Your browser reports for each voice whether a speech engine on your device provides it or a remote speech service. Voices from a remote service are marked online here. The Web Speech specification notes that remote voices may bring extra delay or data use. For private text, pick a voice without the online label.
Why is only the sentence highlighted, not each word?
Word highlighting depends on the voice. The Web Speech specification only asks browsers to report word positions when the speech engine provides them, and some voices do not. With those voices the tool highlights the current sentence instead. Try another voice if you want word by word highlighting.
Is my text uploaded or saved?
Freebiefy never receives your text. It stays in this browser tab and is not saved anywhere. Only your voice and speed settings can be kept on this device, and only if you tick the box for it. A voice marked online is synthesized by a remote speech service rather than on your device, so for private text pick a voice without that label.
How fast can the voice read?
The speed slider goes from half the normal speed to twice the normal speed. The Web Speech specification defines speed as relative to the normal speed of each voice, so 1.5 on one voice is not exactly as fast as 1.5 on another. Pitch goes from 0 to 2 with 1 as the normal pitch.