How to install extra system voices for browser text-to-speech

Why "Loading voices…" sometimes shows only one or two names

The voice dropdown in our converter is populated entirely by your operating system. The browser does not ship voices of its own; it asks the OS, "what TTS voices do you have installed and have permission to use?", and lists whatever comes back. So when the dropdown looks bare, the issue almost always lives one layer below the browser, in your OS's speech settings.

Three things are happening behind the scenes:

• Most operating systems ship with a small default set of voices in the system language only.
• Higher-quality "neural" or "enhanced" voices are usually offered as optional downloads, separate from the base voices.
• Voices loaded after the page opens may not appear until the page is refreshed, because the Web Speech API only re-queries the list on certain events.

Windows 10 and Windows 11

Windows splits voices into two pools: the older SAPI 5 voices used by accessibility tools, and the newer Microsoft "natural" voices that also feed Narrator and Edge Read Aloud. Browsers that use the Web Speech API see the first pool reliably; the natural voices appear in some browsers and not others, depending on how that browser bridges to the Speech Platform.

Adding more languages and voices

1. Open Settings → Time & language → Language & region.
2. Select Add a language and pick the language you want.
3. When prompted for language features, check Text-to-speech and let the package finish downloading.
4. Restart your browser and reload the converter. The new voice should appear in the dropdown grouped under that language.

Enabling the natural / enhanced voices

Search the Start menu for Speech settings. Scroll to Manage voices and use Add voices. On a Windows 11 build with Narrator's natural voices, you can also add them under Accessibility → Narrator → Add natural voices. Browsers vary in whether they pick these up; if a voice you just added is missing from the converter dropdown but works in Narrator, that's a browser limitation, not a Windows one.

macOS

macOS bundles a generous base list and lets you download dozens of additional voices, including high-quality "Enhanced" and "Premium" variants that can be hundreds of megabytes each.

1. Open System Settings (Apple menu → System Settings).
2. Go to Accessibility → Spoken Content → System voice and click the system-voice dropdown.
3. Choose Manage Voices….
4. Tick the languages you care about; for each language, an Enhanced or Premium option is usually available below the standard set. The system downloads them in the background.
5. Quit and reopen Safari, Chrome, or whatever browser you use. The new voices appear in the converter.

Premium voices in macOS are noticeably more lifelike than the standard set, but they take longer to load on a cold start and can stutter on older Macs.

iOS and iPadOS

iOS shares the speech engine with VoiceOver, Spoken Content, and Speak Screen. Adding voices here improves the converter and the accessibility tools at the same time.

1. Open Settings → Accessibility → Spoken Content → Voices.
2. Tap a language and choose a voice. Tap the cloud icon to download an Enhanced or Premium variant.
3. Return to Safari and reload the converter. If the voice still does not appear, swipe up to close Safari fully and reopen it.

Android

Android lets you swap the entire speech engine — Google's, Samsung's, or a third-party engine — and each engine has its own voice catalog.

1. Open Settings and search for Text-to-speech (sometimes under System → Languages & input or Accessibility).
2. Pick your Preferred engine (most users will keep "Speech Services by Google").
3. Tap the gear icon next to the engine, then Install voice data.
4. Select a language and download the higher-quality variant if offered.
5. Force-stop and reopen Chrome.

If you use Samsung's built-in engine, the equivalent screen lives under Settings → General management → Text-to-speech. Switching engines globally can change the default voice across your phone, including for navigation announcements, so try one engine at a time and see what fits.

ChromeOS

ChromeOS exposes both built-in Google voices and downloadable extensions. The built-in set is straightforward:

1. Open Settings → Accessibility → Text-to-Speech.
2. Pick Speech rate and Speech engines. Each engine lists its voices.
3. Visit the Chrome Web Store if you want third-party voices and install an engine extension.
4. Open the converter in a new tab.

Linux

Browser TTS on Linux is the most variable of any platform because there is no single mandated speech daemon. Most desktop distributions ship with speech-dispatcher driving espeak or espeak-ng, which sound robotic but support many languages. For more natural voices, install Festival, Mimic 3, or Piper from your package manager and configure speech-dispatcher to use them as the default output module. After restarting your browser, the new voices should appear in the dropdown.

Common mistakes

• Refreshing instead of restarting. A new voice often does not show up until the browser process is fully restarted, not just the tab.
• Forgetting Enhanced is a separate download. The default voice for a language and its Enhanced version are listed as two separate items; the small one is downloaded by default.
• Installing in the wrong user profile. On Windows, voices added in one user account are not always available to another. Install while signed in to the account that runs the browser.
• Confusing system-language voices with all voices. Some voices only appear if the corresponding system language is also installed. Adding "Italian" purely for TTS still requires the Italian language pack on Windows.

Checking what is actually available

If you want to see the raw list your browser sees, open the converter's voice dropdown — every available voice is grouped by language. If you expected a voice and it is not there, the question is whether the OS exposes it. On macOS you can verify with System Settings → Accessibility → Spoken Content; on Windows, with Settings → Accessibility → Narrator; on iOS and Android, with the Accessibility > Spoken Content menus described above. If the voice exists in the OS but not in the browser, switching browsers usually resolves it — Chromium-based browsers and Safari read the OS speech APIs differently.

What the voice list looks like once you are set up

Once a couple of extra voices are installed, the dropdown groups them by language: English voices grouped together, Spanish grouped together, and so on. Voices marked "(Local)" run on the device itself rather than streaming from the cloud, which matters if you are offline. The converter automatically prefers English voices first, then sorts other languages alphabetically — you can change selection per session, but it is not stored across browsers because we do not require accounts.

Once you have voices you like, the next things to look at are how to write text so it reads well aloud and, if a voice you want is missing on your browser of choice, the browser-support page. For background on the project itself, see the about page.