How to Convert Text to Audio for Free

If you want to turn an article, essay, or block of notes into an audio file without paying, signing up, or uploading anything, freetexttospeech.app does it in about four clicks. Here is the full process, plus what to do when the browser gets in the way.

The short version

Open freetexttospeech.app. Any modern browser works; Chrome or Edge give the best recording support.
Paste your text. Drop it into the editor at the top of the page. No character limit.
Pick a voice. The dropdown lists every voice your OS and browser provide, grouped by language.
Set rate, pitch, volume. Drag the sliders to the sound you want. You can preview first with the Play button.
Click Download audio. freetexttospeech.app plays the speech and captures it to a .webm audio file on your device.

What you get

The output is a .webm file encoded with Opus, sampled at your OS's default voice quality. Every modern player — VLC, QuickTime, Windows Media Player, browser-based podcast apps, Audacity, ffmpeg — reads .webm natively. If your workflow requires .mp3, convert it in any free audio converter, or upgrade to Pro voices on freesuite.app which export .mp3 directly.

Which browsers actually record the audio

Capturing synthesised speech as a file is a quiet corner of the Web Speech API. Support isn't universal yet:

Chrome and Edge (desktop): reliable. The download button produces a real audio file.
Chrome on Android: works on most devices.
Safari (desktop and iOS): playback works, recording is limited. On Safari, freetexttospeech.app will still play the audio aloud — use your device's built-in screen recorder to capture it if needed.
Firefox: similar to Safari. Fine for listening, patchy for recording.

If the download produces a very small or silent file, your browser is one of the restricted ones. Switch to Chrome or Edge and try again.

Tips for a clean audio file

Mute notifications

Because freetexttospeech.app records live audio, any system beep that fires while the recording is in progress ends up in the file. Put your phone on silent, quit apps that might ping you, and start the recording.

Split very long texts

The Web Speech API can stall on texts over ~30,000 characters in Chrome. freetexttospeech.app kicks the engine every 10 seconds to prevent this, but for a full book, split it into chapters and record each one separately. You end up with neat, navigable files anyway.

Use a good voice

The voice matters more than the rate or pitch. On macOS and iOS, the built-in voices (Samantha, Daniel, Karen, Tessa) sound surprisingly natural. On Windows 11, look for voices labelled "Natural" — these are Microsoft's newer neural voices. On Chrome any OS, the Google voices (for example, "Google US English") are a safe bet.

Save a clean .txt copy

Use the Save .txt button in the editor toolbar before recording. That way you have the source text alongside the audio — useful for editing or re-recording later.

When freetexttospeech.app isn't enough

For truly natural audiobook-quality audio, the browser's built-in voices still sound synthetic. Options in that case:

Use the Pro AI Voices upgrade on freesuite.app — neural voices generated server-side, exported as .mp3.
Use macOS's say -o command from the terminal to render an .aiff directly from a voice.
Use a paid service like ElevenLabs, Play.ht, or Amazon Polly for production audio.

For most personal use cases — listening to an article, sharing a quick recording with a friend, adding voice-over to a slide deck — the free browser output is fine.

Open freetexttospeech.app

Paste, pick a voice, download. Free forever.

Start converting →

Frequently asked questions

What audio format does freetexttospeech.app produce?

A .webm file encoded in Opus, which every modern media player and audio editor supports. You can convert it to .mp3 with any free audio converter if needed.

Which browsers can record the audio?

Chrome and Edge are the most reliable. Safari and Firefox have more restrictions on capturing synthesised audio — on those browsers you can still listen but the download may not include real audio. Use Chrome for the best result.

Is there a length limit?

No hard limit. Very long texts (10,000+ words) may need to be split into chunks because some browsers pause long-running speech — freetexttospeech.app works around this with a keep-alive, but it's still smart to split a whole book into chapters.

Do I need to keep the tab open while it records?

Yes. Recording is live — freetexttospeech.app captures the audio as the browser speaks it. Switching tabs or locking the device mid-record will cut off the recording on most systems.

Is the audio watermarked?

No. There are no watermarks, no spoken brand tags, no ads in the audio. You get the raw synthesised speech.

Can I use the audio commercially?

The license on the voices themselves belongs to whoever made them — Apple, Microsoft, Google, or your OS. For commercial audio production, check the voice provider's terms. For personal, educational, or internal use, freetexttospeech.app has no restrictions.