0

I've found and installed this command line text to speech project.

In Windows 10, I have gone to "Speech settings", and under "Installed voice packages", I installed a non-English voice package. Then rebooted, as the instructions said. The new "voice package" is listed in that GUI window, so it's definitely installed.

But voice.exe -l doesn't list it! See:

"Microsoft Hazel Desktop" - Adult,Female,en-GB
"eSpeak-en+f2" - Adult,Male,en-US
"eSpeak-en-us" - Adult,Male,en-US
"eSpeak-en" - Adult,Male,en-US
"Microsoft Zira Desktop" - Adult,Female,en-US

As you can see, there's just the useless "eSpeak" voices (all sound the same, like a robot) and the English Microsoft voices.

The non-English Microsoft voice package is not seen by voice.exe.

All I want is to be able to make my Windows 10 speak to me with a non-robotic voice, on the command line, with my native language. Why doesn't voice.exe see the proper voices? The English voices sound bizarre when trying to speak my language.

PS: I've now spent 15 minutes trying to get StackExchange to recognize the linebreaks in the above output dump. I now give up. I have linebreaks. I can't help if it doesn't care about them.

1 Answer 1

0

Why doesn't voice.exe see the proper voices?

Only a subset of all the Microsoft voices available in Windows are made available system-wide (i.e. to third-party applications). The majority of them are restricted to internal "Windows-only" use. The basic solution to this issue is to modify the registry to make these voices available to other applications.

There are a number of tutorials on the web detailing how to (potentially) unlock these voices for use by third-party software, such as this one on the broad steps for any voice and this one for Eva (Cortana) specifically. However, be aware that Windows 10 is "fluid" and the exact steps may be different for different versions of the operating system.


It's also probably worth mentioning that there could (theoretically) be technical compatibility issues depending on how the third-party software implements its TTS interface (e.g. what API it uses,etc.).


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .