green logo
green logo

Welsh National Language Technologies Portal

NVDA Plug-in

NVDA logo
The NVDA Cymraeg plug-in, developed by the Language Technologies Unit, uses bilingual synthetic voices in order to promote accessibility for Welsh speakers.

This addon provides Welsh neural text-to-speech voices for the NVDA screen reader. It uses Piper TTS models to produce natural-sounding Welsh speech.

NVDA logo

System Requirements

  • Operating System: Windows 10 or 11 (x86 or x64 architecture)
  • ARM64 Windows is not supported (e.g., Surface Pro X, ARM laptops)
  • NVDA Version: 2025.1 or later
  • Internet Connection: Required for initial voice download (approximately 77 MB)

Installation via the NVDA Addon Store

  1. Open the NVDA menu by pressing NVDA+N.
  2. Select Tools, then Addon Store.
  3. In the Addon Store, press Tab until you reach the Channel combo box, and select Beta/dev. This is required because the addon is currently in the beta channel.
  4. Press Tab to the search box and type Welsh Neural Voices.
  5. Use the down arrow to find Uned Technolegau Iaith – Welsh Neural Voices in the results list.
  6. Press Enter to view the addon details, then select the Install button.
  7. Confirm that you want to install the addon when prompted.
  8. When installation is complete, restart NVDA when prompted.
  9. After restarting, a dialog will appear asking if you would like to download Welsh voices (approximately 77 MB). Select Yes to begin the download.
  10. NVDA will announce download progress at 25%, 50%, and 75%. When complete, select OK to restart NVDA and activate the voices.

Selecting the Welsh Voice

After installation, you need to select the Welsh synthesizer:

  1. Press NVDA+N to open the NVDA menu.
  2. Select Preferences, then Settings.
  3. In the Settings dialog, navigate to the Speech category.
  4. In the Synthesizer combo box, select Uned Technolegau Iaith – Welsh Neural Voices.
  5. Press OK to confirm.

Adjusting Settings

You can adjust the Welsh voice through the NVDA speech settings:

  1. Press NVDA+N, select Preferences, then Settings, then the Speech category.
  2. Voice: Use the Voice combo box to choose between 3 speaker variants.
  3. Rate: Adjust the speed of speech.
  4. Volume: Adjust the volume level.

Support

If you encounter any problems or have suggestions, please report them on the GitHub issues page.