The Windows system comes with various voices for all languages installed on your PC. When you install a new language, you have the option to also download the corresponding text-to-speech voices. These TTS voices are available for Cortana, Narrator, Edge Read Aloud and other Microsoft products and service. Unfortunately most of them are not available for third-party software and services. Today, we will show you how to unlock Windows system text-to-speech voices and make them system-wide and available to any third-party programs as well.
Backup Registry
The method requires Registry editing, thus we recommend you to create a backup of the Windows Registry first. In case anything goes wrong, you will be able to restore the initial state. You can directly back it up with the Windows Registry Editor, a built-in program in Windows operating system. If you are on Windows 10, type “regedit” (without quotes) in the search box, then select Registry Editor from the result to open it. Once you are in Windows Registry Editor, click to select Computer from the left panel, then click File menu and choose Export from the drop-down menu. Then you get a pop-up dialog where you can type a name for the backup and choose a location to save it to.
Unlock text-to-speech voices on Windows PC
Launch Windows Registry Editor, browse to Computer\HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Speech\Voices\Tokens, you will see a list of TTS which are available to third-party software on your PC, such as TTS_MS_EN-US_DAVID_11.0, TTS_MS_EN-US_ZIRA_11.0, etc. And you will find the same list of text to speech voices from HKEY_LOCAL_MACHINE\SOFTWARE\WOW6432Node\Microsoft\SPEECH\Voices\Tokens.
To find more TTS that are installed on the Windows system but not currently available for third-party programs, browse to Computer\HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Speech_OneCore\Voices\Tokens. From this list, right-click on a text-to-speech voice you like to unlock and select Export. Then type a name for the new Registry file and save it to your computer hard drive as a .reg file.
Basically what we have to do is to copy a TTS voice from the \Speech_OneCore\ branch to the \Speech\ and \WOW6432Node\ branches as mentioned above. Now right click on this .reg file, and choose Notepad to open it. We can use Notepad or other compatible editors to modify the exported Registry file.
Once the Registry file is opened, you will see it starts with ‘Windows Registry Editor Version 5.00′ as the first line. You will see two locations with or without the \Attributes in the end. These are the places we need to update.
Copy all information except for the first line and paste the information below the original content of the file. That’s to say, you now get two sets of the same data. We now need to replace the location in the first data set with HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Speech\Voices\Tokens; and replace the location in the second data set with HKEY_LOCAL_MACHINE\SOFTWARE\WOW6432Node\Microsoft\SPEECH\Voices\Tokens. And don’t forget each set includes two paths that you need to replace. We have already mentioned above that there are two locations with or without the \Attributes in the end in the original file. Save the change to the exported Registry file in Notepad. And quit the Notepad program on your PC.
It’s time to import the new data into the Windows Registry. Just double-click on the modified Registry file, you will get a warning prompt, and accept the warning you will then see the import successful message which means you can now find the new text-to-speech voice from the \Speech\ and \WOW6432Node\ branches in Windows Registry Editor. The new TTS voice becomes available system-wide now. You can use it in third-party software and services as well. If not, sign off and on again, or restart the PC to allow the change to take effect.
Repeat the above steps to add and unlock extra text-to-speech voices on your Windows computer.
Note that the above method can’t help you unlock Cortana’s voices as they are referenced in another location in the Registry, we will talk about it in another post.