Text to Speech

The Text to Speech Guide

How the reader splits your text into parts, what every control does, and why the voices you hear depend on your browser and operating system.

Open Text to Speech

What the Tool Does

The Text to Speech tool reads text you enter out loud, using the speech synthesis feature built into your web browser. People use it to check a draft by ear, follow a long article while resting their eyes, hear how a sentence sounds before publishing, or practice pronunciation in a language they are learning.

The page is a thin layer over that browser feature. It does three jobs: it divides your text into manageable parts, hands each part to the speech engine with the settings you chose, and tracks where playback is so you can pause, resume or stop. The page has no voice library of its own, and it does not contact a server while it works.

Control by Control

Work through the controls in the order they appear on the tool page. Most of them can be changed at any time, but some only take effect on the next part of the text to be spoken.

  1. Text box. Type or paste up to 50,000 characters. The sentence in the box when the page opens is only a placeholder, so select it and replace it. Each line break starts a new part, so a list with one item per line is read item by item.
  2. Character and word counter. This line updates on every keystroke and counts spaces and punctuation in the character total. Use it to judge how much material you are about to hear.
  3. Limit and long-text notes. Above 5,000 characters, a yellow note explains that playback will run in several parts and may take a while. Above 50,000 characters, a second note says that only the first 50,000 characters will be read, so trim the text or run the rest separately.
  4. Voice picker. This list shows every voice the browser reports, sorted by language code and then by name. The first entry, Browser default voice, lets the browser choose. Entries marked as possibly using an online service are voices whose audio is produced by a provider the browser selects; the page cannot see which provider that is.
  5. Speed slider. Ranges from 0.5x to 2.0x in steps of 0.1. The value 1.0x is normal speed for most voices.
  6. Pitch slider. Ranges from 0 to 2 in steps of 0.1. The value 1.0 keeps the voice's own pitch. Moving far from 1.0 can make some voices sound strained.
  7. Volume slider. Sets loudness from 0% to 100% inside the browser. Your device volume still applies on top of this setting.
  8. Play. Always starts from the first word of the text, even if a run is already in progress. Use it again after you edit the text or change the voice.
  9. Pause and Resume. This button is active while speech is running or paused. Its label changes between Pause and Resume to match the current state.
  10. Stop. Cancels playback and clears the queue of parts. Your text and slider settings are kept, so Play starts again cleanly.
  11. Status line. Reports the part being read, a pause, a stop, a finished run, or an error code if the browser reports one.

Worked Example: How Text Becomes Parts

Consider this sample of 226 characters, which has three sentences and a line break:

The garden needs water on Monday and Thursday. Check the soil with a finger before you pour, and skip the watering if it still feels damp. Tomato plants need the most attention, especially in July. [line break] Remember to rotate the hose.

The tool uses a limit of 180 characters for each part. The first two sentences together are 138 characters, so they fit in one part, and the third sentence cannot join them because that would exceed the limit. The line break then starts a fourth piece. The result is three parts:

PartCharactersContent
1138The first two sentences
258The sentence about tomato plants
328The line about the hose

The status line reads "Reading part 1 of 3" first, then moves to part 2 and part 3. If you change the speed to 1.5x after part 1 has started, part 1 keeps its original speed and parts 2 and 3 use 1.5x. Splitting by sentence keeps the reading natural, while the size limit keeps each piece short enough for browsers to speak reliably.

Tips and Common Mistakes

  • Audition the voice first. Paste one paragraph and listen before committing to a long run. Voice quality differs a great deal between voices and between browsers.
  • Clean up formatting. Symbols such as asterisks, hash marks or bullet characters may be read aloud, skipped, or read differently depending on the voice. Remove them from the text when they are not meant to be heard.
  • Expect short gaps. A slight pause between parts is normal, because each part is handed to the browser separately.
  • Do not expect instant changes. Moving a slider during a run changes the next part, not the one already speaking. Press Play to hear every setting from the beginning.
  • Do not expect every installed voice. Voices that a browser does not expose will never appear in the list, no matter how many voices your operating system has.
  • Keep the tab active on phones. Some mobile systems stop speech when the screen locks or the browser moves to the background. Keep the screen on for long readings.

Voices and Privacy

Voice availability is decided by the browser and the operating system. Some voices are local to the device, while others may be produced by an online service the browser selects. This tool cannot tell which kind of voice it is using, except for the label shown in the list. The terms and data handling of any online service are set by its provider, and you can review the browser's own settings to see what it allows.

What this page does is limited. It does not send your text to a server, it does not write the text to browser storage, and it does not record the audio. The text lives in the open page only, so closing the tab removes it.

Frequently Asked Questions

Why does the same text sound different on another computer?

Each computer and browser exposes its own set of voices. Two machines can have different voice packs installed, different operating system versions, or different browsers, so the same text can be read by completely different voices.

Is this tool a substitute for an assistive technology product?

It can be a helpful way to hear text, but it is not a certified assistive technology product. If you rely on a screen reader or a read-aloud feature for access, test this tool with the software you already use before depending on it.

Can I save the spoken audio as a file?

No. The tool plays speech through your device and does not record, export or download the audio. Recording system audio is a separate job that needs other software.

Why did playback stop on its own?

A browser may end speech when the tab moves to the background, the device sleeps, or the voice service fails. When the browser reports an error code, the status line shows it, so you can see whether the cause was an error or a normal stop.

Open Text to Speech