Sound device issues.

[Gtrshop]

I use a MOTU 828 firewire device for my audio. This device is a recording interface that gives me multiple ins and outs, and allows me to send the audio to different stereo pairs.

Generally this works.

Lately I seem to be having an issue with VA using the correct device for output.
I have 4 devices listed for output by name:

MOTU 3/4 (Main, gaming) (this would be system default device)
MOTU 5/6 (TS/Discord)
MOTU 7/8 (VA/SRA)
MOTU 1/2 (MAINS)

As you can see Voice attack is supposed to be coming out of MOTU 7/8, which is NOT the default device. When I first start up my computer and open VA it is defaulting to the MOTU 3/4 (Main, gaming). I go into to the AUDIO tab and Override default playback device is set to MOTU 7/8 (VA/SRA), where I want it, as is Override default txt to speech device. VA is still playing through the MOTU 3/4 (Main, gaming).

I see the option to set the Windows audio dive on startup, and that is NOT selected, although i could. When I select that, VA does come out of the correct speaker, but the reassignment messes up another piece of software. So, I'd rather not use the Set Windows Default on startup if possible.

Am I missing something to make VA come out of one device without it having to be the default device?

[Pfeil]

Do you have the "Audio Output Type" option set to "Integrated Components"?

[Gtrshop]

It was set to Windows Media Components.  I've changed it.

[Gtrshop]

So far everything seems to be good, the issue wont show up until I got through my routine tomorrow, then I'll know for certain...

[Gtrshop]

So, yes, that did setting down the device assignment issue.

Since setting to integrated components, I seem to get V/A talking over itself. I'm guessing there are multiple informations that are going to be spoken, but more than 1 will be spoken at the same time, makes it horribly confusion to listen to.

Single TTS Instance is selected and always has been.

[Pfeil]

Make sure you have the "Wait until speech completes before continuing command" and/or "This completes all other currently-playing text-to-speech" options checked in "Say Something with Text-To-Speech" actions you don't want to play simultaneously.

"Single TTS Instance" only means the TTS engine will render audio one at a time(serially using single instances, rather than running multiple instances in parallel) not that it will play one at a time.

[Gtrshop]

So check both of those to be enabled? And, is that something I have to do PER COMMAND?

[Gtrshop]

There are 651 commands to go though.... I looked inside a couple, several have two or 3 separate "Say" commands that need to have those boxes ticked off. Something doesn't seem right about that...I didn't have to select them before,and the commands  spoke only 1 at a time.. I'm not sure what changed to cause that.

I did check them - those items - off in a couple commands though, but now, its not saying enough. It should be telling me I have a "new fast lap' and nothing is being said.

[Pfeil]

"Integrated Components" allows multiple TTS phrases to be spoken at the same time(as it caches the output of the TTS engine and plays it back using its own sound system).

If the profile was created without this in mind, instead relying on what is effectively a limitation of the TTS system to not have audio playing simultaneously, that could cause what you're experiencing.

[Gtrshop]

That makes sense. I didn't create the profile, it comes with the software I've bought. There are 3 sources of commands amalgamanted into one. Is there a log of what commands get utilized? What Im thinking is the log might point me to the source areas to take a look at. It could be one supposed of commands and not all of them, that would narrow things down considerably.

[Pfeil]

The only built-in log that will display commands is the one on VoiceAttack's main window. Unless the profile(or your VoiceAttack settings) hides them explicitly that should list which commands have been executed during that session.

Do also make sure that the third-party software(and/or VoiceAttack plugin) doesn't use TTS directly, alongside VoiceAttack.

[Egil Sandfeld]

Hi there. I'm the developer of the 3rd party plugin for VA here. I'll make sure to not rely on the single TTS instance, and in a way I already am. I just think there are some gaps here and there.

Thanks for digging into this!