Using non-default audio devices?

[Exigeous]

I have 7 soundcards (long story why) and I need Voice Attack to use a different soundcard for the mic input than the Windows default. It seems I can change this when I launch it but then it doesn't keep those settings so I have to remember to change it every single time I launch the app. I do NOT want it to change my Windows default devices (which it seems to do constantly even with the check boxes unchecked).

I've tried setting "Default Multimedia Audio Recording Device" then click "Change Now" with the check box UNCHEKED - I've then changed "Override Default Text-to-Speech Device" but that doesn't seem to change the device it's actually listening on as it's still listening to the wrong mic. I also can't get it to launch with the mic muted as if I close it with it muted it reopens unmuted again.

Am I missing something or is this just not possible as it's VERY frustrating.

Thanks!

~Exigeous

[Pfeil]

To change the device the speech recognition engine uses you can use the "Windows Speech Recording Device" option on the "Recognition" tab of the VoiceAttack options window.

Note that this option applies to all applications using the Microsoft speech recognition engine on your machine.

[Exigeous]

This appears to be incorrect - as if I change that to a audio device that doesn't actually have a microphone VA is still "hearing" me and translating what I say in the main window. It seems the only way is to change the devices below and hit "Change Now" which I have to do on every single launch.

So what is the difference between the two?

[Pfeil]

That option is a shortcut to the one described in the last paragraph of this topic; After changing it from the VoiceAttack options window and clicking "OK", there should be a "Recording device changed" entry in the log on VoiceAttack's main window, at which point the speech recognition should be listening to the chosen device.

It sets up the speech recognition engine to use a specific device, rather than the default multimedia recording device (which is one of the default devices that can be changed on the "Audio" tab).

[Exigeous]

Can confirm that is NOT what I'm experiencing. There is no message in the app when I change the device and regardless of what that is set to it is still using my default Windows mic. So to restate:

Devices
- Sound 1 (primary XLR<>USB mic)
- Sound 2

If I set that to "Sound 2" there is no indication in the app, however when I restart the app it does stay set to "Sound 2" - however "Sound 2" has no working microphone but I am still getting speech to text in the app as it is still using my default Windows device of "Sound 1".

Maybe I can go at this another way - my goal is this, I NEVER want to use VA for voice input. NEVER. I only use it as a macro tool that is triggered by my joystick devices and/or keyboard inputs. If I "mute" the keyboard input in VA that setting persists between restarts however the "headset" icon does not persist, so I have to remember to click it on each launch and often times something re-enables it (and I can't determine what). Even when that is clicked I see it is still "listening" based on the volume meter shown at the bottom of the app.

I appreciate the help, not sure why it's working differently for me. I am running it as administrator BTW.

~X

[Pfeil]

If you don't plan on using speech recognition, you could disable that functionality by checking the "Disable Speech Recognition" option on the "Recognition" tab of the VoiceAttack options window (right next to the "Windows Speech Recording Device" dropdown).

[Gary]

The Windows speech engine can be set to a specific recording device, or it can be set to whatever the user has set up as the default recording device.  Either way, it's a single connection point - one is constant, the other can float.

In VA's settings, there's an option that lets you set Windows' default recording device on startup (audio tab) - this is the older option.  This relies on the user leaving the speech engine set to 'default audio device' within its control panel settings for speech recognition to work.  It's an awkward option for sure - it serves a dual purpose (it rode in with some other features - it's why it is on the audio tab).  I can't remove it, as there are folks that use it for one reason or another.

Later on, the 'Windows Speech Recording Device' option was added to the recognition tab.  This is a shortcut to changing the specific recording device to use for the speech engine across all applications.  This relies on the selected device to always be present and enabled - if it's not, no speech recognition will occur.  It changes a system setting that is not normally used (or even known about) - I hesitated to add it for the longest time.

If your setup has the Windows speech engine pointing to a specific device (either set up through VA or through control panel), and there is audio coming from other sources that is getting into that channel, you've got some type of bleed going on.  This is going to need to be mitigated locally, unfortunately, as this indicates either a hardware/driver issue or some type of software routing configuration.   As Pfeil had mentioned, turning off speech facilities will definitely eliminate the problem of speech recognition occurring during your VA session no matter what channels are used

Hope some of that helps clear up some confusion!

[Exigeous]

D'oh - how the crap I missed the option to just disable it is beyond me - should have started with my problem rather than the solution!

Thanks, have that set and all good now!