A text to speech app reads written text aloud; this browser-based option accepts up to 20,000 UTF-16 code units and uses a voice supplied by your browser and operating system.
For a straightforward browser workflow, the Lizely Text To Speech tool lets you enter plain text, choose from the voices the device reports, adjust rate and pitch, and control playback without creating a separate audio file.
The text stays in the current tab and is passed only to the browser speech interface; Lizely does not upload or store it. That keeps your writing from being sent to Lizely, but it is not a blanket promise of offline availability because browser support and the speech engine still depend on the device you are using.

What the text to speech app includes
This is a focused browser utility built around plain text and direct playback. It does not expose a server-side voice. Instead, the browser and operating system supply the voice, while the tool manages the text, chosen settings, playback queue, and status.
| Part | Supported behavior | Practical meaning |
|---|---|---|
| Input | Type or paste plain text, up to 20,000 UTF-16 code units. | Blank, null, and oversized input is rejected. |
| Voice | Select from the voices reported by the current browser and operating system. | Names, languages, accents, and availability can vary by device. |
| Rate | Adjust from 0.5× through 2.0×. | It is a relative browser control, not a measured words-per-minute value. |
| Pitch | Adjust from 0.5 through 2.0. | Different speech engines interpret the value differently. |
| Playback | Speak, Pause, Resume, and Stop are explicit controls. | Editing the text or a setting stops the active queue. |
| Output | Live browser playback only; no audio file is synthesized. | The result cannot be exported as MP3, WAV, or another recording. |
| Data handling | The text remains in the current tab and is passed to the browser speech interface. | Lizely does not upload or store the text. |
The accepted material is plain text. Line endings and repeated horizontal spaces are normalized before playback, while words and punctuation are not rewritten.
Read plain text aloud in the browser
- Open the Text To Speech tool and enter your text. Type or paste the passage you want to hear, using plain text and staying within the 20,000 UTF-16 code-unit limit. Blank text, null characters, and oversized input are rejected.
- Select an available voice when one is offered. Voice choices come from the current platform. If no explicit voice list appears, the browser has not exposed additional choices, and the tool cannot install extra operating-system voices.
- Adjust rate or pitch if needed. Rate accepts values from 0.5× to 2.0×, while pitch accepts values from 0.5 to 2.0. These are relative controls rather than measured speech speed, musical pitch, or accessibility certifications.
- Select Speak. This cancels any older queue created by the widget and queues the current text chunks in order for the browser speech engine.
- Manage the reading with the playback controls. Use Pause, Resume, or Stop at any point. Editing the text or a setting stops the active queue, allowing you to make a change and select Speak again for a fresh reading.
The browser may require a user gesture before audio output, so begin with the visible Speak control. If it rejects a selected voice or encounters another playback failure, the tool reports a playback error rather than claiming that the speech completed.
Choosing a voice, rate, and pitch
Voice names, languages, accents, quality, pronunciation, and availability come from the current browser and operating system. A voice visible on one computer may be absent on another because each platform exposes a different set of speech engines. The tool does not download additional voices. The browser-facing foundation is described in MDN’s SpeechSynthesis reference.
Rate and pitch are applied to every queued chunk. They are SpeechSynthesisUtterance control values, not measured words per minute, musical intervals, or accessibility certifications. Different engines interpret them differently, so a setting that sounds suitable on one device may need adjustment on another. MDN’s SpeechSynthesisUtterance documentation explains the underlying browser controls.
Pause and Resume are delegated to the same browser engine. Because that engine controls the global speech-synthesis queue for the current page, another script using the same browser API could interact with playback. Stop and edits cancel this widget’s queued speech, and unmounting the widget also cancels it.
How long passages are handled
For long input, the tool does not send one huge utterance to the browser. It divides the text into bounded chunks of no more than 220 UTF-16 code units. The splitter prefers a sentence boundary, then a clause boundary, then a space, and finally applies a hard boundary to an unusually long unbroken token. This avoids sending one very large utterance to an engine that might stop or behave inconsistently on lengthy input.
The split is mechanical rather than linguistic. The tool does not provide linguistic sentence parsing or promise identical pauses across voices. It also does not rewrite your words or punctuation. Only line endings and repeated horizontal spaces are normalized before the chunks are sent to the browser.
Limits that matter before you rely on it
A browser speech app can cover a quick listening task without becoming a complete audio-production package. Understanding these boundaries prevents you from searching for an export or control the tool does not provide.
- No downloadable recording. The standardized browser speech interface exposes playback controls but not a portable audio buffer, so the output cannot be saved as MP3 or WAV.
- No screen reader replacement. The tool reads entered text only. It does not announce page structure, focus, controls, semantic roles, or live regions.
- No fixed pronunciation dictionary or SSML workflow. Use a dedicated synthesis service when repeatable pronunciation rules or tagged markup are required.
- No commercial voice library. The tool uses voices reported by the local platform rather than offering a catalog of licensed commercial voices. For that requirement, use a dedicated service and review its privacy and licensing terms.
- No substitute for essential assistive technology. For essential assistive use, rely on a maintained screen reader and your platform’s accessibility settings.
The tool is also not a medical, diagnostic, language-learning, transcription, voice-cloning, or accessibility-conformance service. It cannot guarantee offline availability, and browser or operating-system support remains outside its control.
Tasks this browser app fits
Use the app for work that matches live, system-voice playback. The table separates suitable listening jobs from requirements that need a dedicated synthesis or recording service.
| Task | Fit | Important detail |
|---|---|---|
| Listen to a short passage | Good fit | Enter plain text and use a voice reported by your device. |
| Review how a passage sounds | Good fit | Rate and pitch can help with rhythm and pronunciation checks. |
| Read a long draft | Within the input limit | The text can contain up to 20,000 code units and is divided into bounded browser utterances. |
| Produce a reusable voiceover | Not a fit | There is no MP3, WAV, or portable audio export. |
| Match a licensed commercial voice | Not a fit | The app uses locally reported voices and cannot guarantee the same voice across machines. |
| Provide essential screen-reader access | Not a fit | It does not convey page semantics, focus changes, or live-region updates. |
A practical listening routine
For a dependable pass, prepare the input before starting playback and keep the controls within easy reach.
- Paste or type the plain text, then confirm that it is within the 20,000-code-unit limit.
- Choose a reported voice before starting playback, especially when multiple voices are available.
- Set the rate and pitch to suit the passage rather than assuming the values will behave identically on another device.
- Test a short section first. Editing the text stops the active queue, so you can return to the full passage and select Speak again.
- If a voice is missing or playback fails, check what the current browser and operating system report. Extra system voices are outside this tool’s scope.
For drafts, notes, short passages, and a quick check of how a system voice handles your wording, this focused workflow covers the job. For a reusable recording, licensed commercial voice, fixed pronunciation rules, SSML, or dependable cross-machine reproduction, choose a dedicated service and check its privacy and licensing terms.
If you're weighing options, Break Reminder Without App: Browser Tab Guide covers this in detail.
If you're weighing options, Online Stopwatch Without App: Browser Timing Guide covers this in detail.