Use multiple Text to Speech voices with VA

[kenjiro]

I have many TTS voices installed on my Windows 10. They are about 10 from Microsoft and 2 from IVONA.

I play Elite Dangerous and I am creating a VA profile with multicrew. That is, I want to call one of those voices and tell it to take care of an "area" of the ship (navigation, engineering, weapons, etc).

By what I read on VA's manual/guide, I should be able to use the "Say something with Text-to-speech" function, then set the "Voice" to a variable name, and then VA would speak to me whatever is in the text box (could be another variable).

So I tried setting the "Voice" to {TXT:Voice} and the text to be read is {TXT:Script}.

If I manually choose a Voice (let's say IVONA 2 Amy), it works. If I type that same string (IVONA 2 Amy) in the Voice section, it works. But if I type {TXT:Voice} there, VA does nothing.

And I even tried making VA write the content of those variables to the log. Both variables had the correct contents.

What am I doing wrong?

Thanks for your attention

[Pfeil]

Seems to work fine on my machine using those variables/tokens and an IVONA voice. Can you post the contents of your command?

[kenjiro]

Thanks for replying.

I attached three images showing a simple command to test what I want

EDIT: The "Voice" variable is set by commands related to the name of the TTS voices I have installed. So I say "Amy", then the command sets the text variable 'Voice' with 'IVONA 2 Amy' (which is the name of one of the TTS Voices my Voice Attack can use)

[Pfeil]

If I manually choose a Voice (let's say IVONA 2 Amy), it works. If I type that same string (IVONA 2 Amy) in the Voice section, it works. But if I type {TXT:Voice} there, VA does nothing.

This is strange, as I've tested with both v1.6.9.7 and v1.6.9, and both use the system default voice when the "Voice" field is set to something that doesn't match a known voice.

Which version are you using? Can you post the command that sets the voice?

[kenjiro]

I am running VA 1.6.9 (bought on Steam)

The command that sets the "Voice" variable is the one attached.

So, nothing fancy. I set two variables with the same content. But for the test I have been doing, I used only the "Voice" variable.

[Pfeil]

Everything looks to be setup correctly, sadly it works fine for me with VoiceAttack Steam v1.6.9 as well.

This is what has me thinking something lower down is going wrong:

If, ‘Default’ is selected as the, ‘Voice’ value (or, if a token or variable name does not resolve to a valid voice name as indicated above), the text-to-speech voice selected in Windows’ Control Panel will be used.

If you type gibberish into the "Voice" dropdown, does the system default voice speak your text?


I will note I'm running Windows 7, which could theoretically make a difference.

[kenjiro]

I did exactly that for a new test. I typed 'gibberish' in the Voice dropdown and the result was the same: VA says nothing.

I even tried type just "Hello world" in the text field, instead of using the value of a variable.

I am using Windows 10 Pro and I have set 'Microsoft Catherine' as Windows' default TTS voice.

Oh, another symptom I forgot to 'repport': when VA does nothing on those tests, it kindda freezes after some time. It looks like the command is still trying to finish.

[kenjiro]

Pfeil, really thanks for all the attention you gave me.

I just tried the very same commands on my other computer (also running Win10 Pro) and they worked. So there is some sort of black magick around my main PC

Now my Voice Attack profile (for those playing Elite Dangerous) can either use VA's default TTS engine or EDDI's engine. Oh and it supports 23 voices   

[kenjiro]

More than 1 month later and I still couldn't figure out what is wrong with my Voice Attack.

Today I even uninstalled it, removed all the VA related files and folders. Then I opened steam and told it to install VA again; then I loaded my profile, told it to use VA for TTS (instead of EDDI). When I issued a command... again VA hanged.

Damn!

Any tips on what I should check to find out what's is going on?

Just to refresh our memories, the problem shows up when I place a variable in the "Voice" field, instead of specifically choosing one.