Press Enter or Space to start voice dictation. Your browser may ask for microphone permission.
Listening. Press stop when you are finished.
iOS / Safari limitation: Natural (AI) voice playback is still restricted on iPhone. For the best experience, use Quick (device) voices or a desktop browser.
AI generation runs in your browser. First use downloads the voice model (saved for future visits).
Natural voices need about 380 MB
Natural (AI) voices run entirely on your device. Before first use, the voice model downloads once (about 380 MB) and is saved for next time.
Natural voice models may fail on iPhone, iPad, and other low-memory devices. This device may run out of memory before a voice is ready. A desktop browser is more reliable, and Quick (device) voices still play here without the model download.
How it works
Paste, type or dictate unlimited text and hit Speak. There is no character or word limit🥳. Device mode uses your built-in voices, while AI mode delivers more natural quality and fast generation. Use the AI download button to save your audio instantly.
You can pause, skip backward/forward, and seek by dragging the progress bar. Most device voices support adjustable speed, pitch, and volume.
Voices & Languages
Theoretical maximum: 100 + languages and 300 + voices if every OS pack, browser-provided online voice pack, and third-party engine is installed.
Typical setups expose 40-60 languages and 120-200 voices. Check the dropdown above to see what your device supports.
Why does the number change?
speechSynthesis.getVoices() displays voices that your browser can access. Install extra voice packs, refresh this page, and they appear automatically.
Privacy first
Where does your text go?
Privacy by mode — what stays local and what may be processed online:
| Mode | Processing | Privacy |
|---|---|---|
| Device voices (Quick) | Web Speech API | Fully local |
| Browser-supplied online voices (if available) | Browser/provider cloud service | Text may be sent to provider servers |
| AI mode (Natural) | Supertonic TTS-3 in browser (WebGPU/WASM) | Fully local after model download |
- Device voices (Quick): fully local via Web Speech API.
- Some browsers may provide online voices; using those may send text to provider servers.
- AI mode (Natural): processes text directly in your browser; fully local after model download.
- Audio is streamed straight to your speakers and never stored by us.
- Open-source — audit the code on GitLab.
Add more voices on your device
Windows 10/11
- ⚙️ Settings → Time & Language → Language & region.
- Click Add a language → Install Speech.
- For premium "natural" voices: Settings → Accessibility → Narrator → Add natural voices.
macOS 12 +
- System Settings → Accessibility → Spoken Content → System Voice.
- Click Manage Voices…, tick extra voices, then Download.
iOS / iPadOS
- Settings → Accessibility → Spoken Content/read & speak → Voices.
- Select a language, tap the ⬇ icon beside any additional voice to download, make sure to pick a premium or advanced voice to get the best sound quality.
Android
- Settings → Accessibility → Text‑to‑speech output (wording may differ).
- Tap the gear next to Preferred engine → Install voice data.
Browser online voices (if available)
Some browsers may expose additional online voices depending on version, platform, account, and region.
Frequently asked questions
How do I install new text-to-speech voices on Windows 10/11?
- Go to Settings → Time & Language → Language & region.
- Click Add a language and install speech for your chosen language.
- For premium "natural" voices: Settings → Accessibility → Narrator → Add natural voices.
How do I add more voices on macOS 12 or later?
- Open System Settings → Accessibility → Spoken Content → System Voice.
- Click Manage Voices…, tick extra voices, then click Download.
How do I install new voices on iOS or iPadOS?
- Go to Settings → Accessibility → Spoken Content → Voices.
- Select a language, then tap the ⬇ icon beside any additional voice to download.
How do I add more voices on Android?
- Go to Settings → Accessibility → Text-to-speech output (wording may differ).
- Tap the gear next to Preferred engine and select Install voice data.