About Text to Speech
Convert text to speech using your browser. Multiple voices, adjustable speed, free. Free, no sign-up required.
How to use
- Paste or type your text into the textarea. The tool sends the entire string to the browser's built-in speech engine, so very long passages work but may take several seconds before audio starts on slower machines.
- Open the Voice dropdown to see every voice your browser and operating system provide. The list is auto-filtered to default to your first English voice; switch to system voices in other languages to test pronunciation or generate language-learning audio.
- Drag the Speed slider between 0.5x (deliberate, useful for accessibility and language learners) and 2x (fast scan, useful for proofreading by ear). 1x is the natural pace; 1.25x is the most common preferred listening speed for podcasts.
- Click Speak to start playback. The status line below shows '🔊 Speaking…' while audio plays, '✅ Done' when finished, or '❌ Error' if the engine fails. Speech runs through your default system audio output.
- Use Pause to suspend mid-sentence and Stop to cancel and reset. Pause and resume are not perfectly seamless on every browser — Chrome handles it well, Safari sometimes restarts the current sentence.
- For proofreading, listen with eyes closed. Hearing your own writing aloud surfaces awkward phrasing, repeated words, and missing connectives that visual reading skips. Many professional editors run every draft through TTS as a final pass.
- To save audio as a file: the Web Speech API does not directly export to MP3 or WAV. Workaround — use your operating system's audio recording (QuickTime on macOS, Stereo Mix or Voicemeeter on Windows) to capture the speaker output while the tool plays.
Frequently asked questions
Why are different voices available on different computers?
The Web Speech API hands voice synthesis off to the operating system. Windows ships Microsoft voices (David, Zira, Mark, Hazel) plus any installed language packs. macOS ships Apple voices (Samantha, Daniel, Karen, Tessa, plus Premium and Enhanced variants if downloaded via System Settings). Chrome OS provides Google voices. Mobile browsers expose iOS Voice Over voices or Android TTS engines. So the same web page running on two different computers will list completely different voice options — that is by design, not a tool limitation.
Can I install higher-quality voices?
Yes. On macOS, open System Settings > Accessibility > Spoken Content > System Voice and click 'Customize' to download Premium and Enhanced voices (some are several hundred megabytes but sound dramatically better). On Windows, install language packs via Settings > Time & Language > Speech. After installing, restart the browser and the new voices appear in the dropdown. Premium Apple voices and the Microsoft Natural voices are nearly indistinguishable from human narration.
Does this support SSML markup?
The Web Speech API has limited SSML support — most browsers ignore SSML tags and read the markup as plain text. For controlling pause length, emphasis, phonetic pronunciation, and prosody, you need a server-side TTS service like Amazon Polly, Google Cloud TTS, Azure Speech, or ElevenLabs which fully support SSML. Web Speech is best for quick, casual playback; SSML-grade applications (audiobooks, IVR, accessibility products) typically require paid cloud TTS.
Why does playback sometimes cut off mid-sentence?
Chrome enforces a hidden timeout that stops Web Speech playback at roughly 15 seconds per call. The official workaround is to split long text into shorter chunks and queue them individually — production TTS apps split on sentence boundaries every 100-200 characters and chain the utterances with onend handlers. This tool does not currently chunk, so single very long passages may truncate. Workaround: paste shorter sections at a time.
Is it useful for accessibility?
Web Speech is helpful but not a substitute for a real screen reader. Screen readers (NVDA, JAWS, VoiceOver, Narrator) understand DOM structure, headings, ARIA landmarks, focused elements, and announce changes intelligently. This tool just speaks whatever raw text you paste in. For accessibility audits, run your content through a screen reader, not a TTS tool. For non-disabled users who want their inbox or articles read aloud while doing other tasks, Web Speech is fine.
Does the tool send my text to a server?
No. Speech synthesis happens locally via the Web Speech API, which routes the text to your operating system's bundled TTS engine. No network requests are made for the audio generation itself. This is genuinely private — you can safely paste sensitive emails, draft contracts, or personal journal entries to hear them read aloud without data leaving your machine.
Why does the same voice sound different in Chrome vs Safari?
Even though both browsers expose the OS voice library, each browser tweaks pitch defaults, applies different audio compression on output, and may use slightly different versions of the underlying engine. Safari on Mac uses Apple's speech engine directly with minimal transformation. Chrome adds light processing. The difference is usually subtle but noticeable on long passages. For consistent audio across browsers, pick the same named voice and rate in each.
Part of ToolFluency’s library of free online tools for Productivity. No account needed, no data leaves your device.