HCS Voicepacks Issue - Sorted and UPDATED

[Para Handy]

OP, consider using Voicemeeter Banana (https://www.vb-audio.com/Voicemeeter/banana.htm ).

It's free (You can pay if you want), it works, as you can set Mic sensitivity and set a Microphone 'gate' up to only let in sounds you want, and there is a whole heap of setup help on YT. It is quite easy to setup once you get your head round what the settings are

Thanks for that - I'll have a look and see how I get on.

I think this is all I did, use at your own risk.

First I set up a Global Hotkey.

-----In voice attack options (wrench icon) select Hotkey tab.

----------Recognition Global hotkey, click on change box to open Global Hotkey window.

--------------- Check box for enable global hotkey.

--------------- Set a key or key combination for the global hotkey (I use LALT F12)

---------------Select Voice Attack listens while keys are down option.

----------------Click OK to close window.

----------Click OK to close option window.

Then edit the Singularity profile.

-----In the Voice Attack window, click on the pencil icon beside the profile name.

----------In the Edit a Profile widow, click on the check box beside the profile name.

---------------In the Profile Options window select the Profile General tab.

--------------------Check the box for “Override listening if my spoken command begins with:”

-------------------------Type the word or phrase you want to use (I use computer).

-------------------------Click Ok to close the window.

Apply change and restart voice attack.

Now that looks really interesting, I shall have to spend time trying it out. Thanks very much for taking the time to answer.

And thanks to all. [up]

 

[CMDR Karrde Sun]

Since noones actually answered this question, im somewhat sure the command is 'interaction mode off/on'

Another way is to say 'protocol override customize my settings' and look for a native windows dialog box voice attack will spawn. Its in there somewhere as well.

Ps. the fairly annoying action prompts are turned off by 'invoke mode off', though you can set which ones you may not want in the protocol override box.
pps. the absolutely amazing 'exploration mode on' you can't live without once you see it work (discovery scanner must be on the secondary fire trigger).

 

[Para Handy]

Since noones actually answered this question, im somewhat sure the command is 'interaction mode off/on'

Another way is to say 'protocol override customize my settings' and look for a native windows dialog box voice attack will spawn. Its in there somewhere as well.

Ps. the fairly annoying action prompts are turned off by 'invoke mode off', though you can set which ones you may not want in the protocol override box.
pps. the absolutely amazing 'exploration mode on' you can't live without once you see it work (discovery scanner must be on the secondary fire trigger).

Thanks for your response. I am still coming to grips with this stuff but...

From the HCS manual:

Invoke interaction
Interaction mode on. Starts the randomised response mode.
Interaction mode off. Stops the randomised response mode.
Some of the virtual assistants have “Interaction mode on/off”
and some of them don’t. With “Interaction mode on” , the
virtual assistant will randomly speak out, ask a question, or
invoke a game of “I spy”. This mode is experimental and in
selected packs only.
Astra will say “I’ve got something on my mind” or something
similar - Your response can be just to ignore it, or, use the terms:
“Okay go ahead” -”Yes, I’m listening” - “Yes Astra, go ahead”
Astra will then respond with some information


So that is not the same as stopping the crew responding to each other.

The customiser does allow modifying these but again that is not the issue.

 

[disorganise]

What baffles me, to be honest, is why it's even an issue to begin with.
I have noise cancelling headphone that (I think) work by 'listening' to the surrounds and playing back an inverted wave.
In contrast the PC 'knows' *precisely* the sound it is making, so ought to be able to completely ignore it. What remains would be your voice, some sound reflections and the ambient.

 

[CMDR Karrde Sun]

Thanks for your response. I am still coming to grips with this stuff but...

From the HCS manual:

Invoke interaction
Interaction mode on. Starts the randomised response mode.
Interaction mode off. Stops the randomised response mode.
Some of the virtual assistants have “Interaction mode on/off”
and some of them don’t. With “Interaction mode on” , the
virtual assistant will randomly speak out, ask a question, or
invoke a game of “I spy”. This mode is experimental and in
selected packs only.
Astra will say “I’ve got something on my mind” or something
similar - Your response can be just to ignore it, or, use the terms:
“Okay go ahead” -”Yes, I’m listening” - “Yes Astra, go ahead”
Astra will then respond with some information


So that is not the same as stopping the crew responding to each other.

The customiser does allow modifying these but again that is not the issue.

Im pretty sure interaction mode does. I had the opposite problem, my crew were not saying random thoughts, and i used the inverse command to turn it on!

EDIT: As an extreme solution, im pretty sure theres a completely silent mode somewhere if you just want the automation.

Also while that section looks correct, the manual is known to be extremely out of date. At the least what you pasted is missing the other 'mode' features If you haven't already tried it, id strongly recommend running 'protocol override customize my settings' and alt-tabbing out to find the configuration dialog window. Their customer support is extremely friendly as well, usually the owner responds and in prompt time, so you could could try the support email as well.

Final tip: Theres an unadvertised set of html cheat sheets that is included in the directory the voice pack installs itself in the voice attack folder. It starts with a '.' Though im starting to suspect those are out of date too...

 

[Para Handy]

What baffles me, to be honest, is why it's even an issue to begin with.
I have noise cancelling headphone that (I think) work by 'listening' to the surrounds and playing back an inverted wave.
In contrast the PC 'knows' *precisely* the sound it is making, so ought to be able to completely ignore it. What remains would be your voice, some sound reflections and the ambient.

I have good headphones with excellent noise-cancelling mic but as I said I don't want to use headphones rather than my 5.1 surround system and sitting with the phones on my head / round my neck just to use the mic is a bit of a pain. It is not an issue at all when using headphones rather than the sound-system.

 

[Para Handy]

Im pretty sure interaction mode does. I had the opposite problem, my crew were not saying random thoughts, and i used the inverse command to turn it on!

EDIT: As an extreme solution, im pretty sure theres a completely silent mode somewhere if you just want the automation.

Also while that section looks correct, the manual is known to be extremely out of date. At the least what you pasted is missing the other 'mode' features If you haven't already tried it, id strongly recommend running 'protocol override customize my settings' and alt-tabbing out to find the configuration dialog window. Their customer support is extremely friendly as well, usually the owner responds and in prompt time, so you could could try the support email as well.

Yea it is not the random stuff that is the issue, it is VA hearing the crew responses and the voicpack misinterpreting that and acting upon it. So far I have had them raise and lower the gear as they think fit, deploy / retract cargo scoop, distribute power, assume attack profiles, etc - all in response to the crew's utterances.

I have indeed used the "Customiser" window that you refer to (see my earlier response) and as I said no it doesn't offer a solution.

EDIT - I don't think I would be happy spending money on a few voicepacks and not have the voices.

P.S. I forgot tot use the multi-quote again - sorry.

 

[CMDR Karrde Sun]

Yea it is not the random stuff that is the issue, it is VA hearing the crew responses and the voicpack misinterpreting that and acting upon it. So far I have had them raise and lower the gear as they think fit, deploy / retract cargo scoop, distribute power, assume attack profiles, etc - all in response to the crew's utterances.

In that case the only thing i could think of is to do additional voice training in the windows speech recognition control panel. I've never had that problem and did the training a few times.. best of luck.

 

[disorganise]

Im pretty sure interaction mode does. I had the opposite problem, my crew were not saying random thoughts, and i used the inverse command to turn it on!

EDIT: As an extreme solution, im pretty sure theres a completely silent mode somewhere if you just want the automation.

Also while that section looks correct, the manual is known to be extremely out of date. At the least its missing the other 'mode' features If you haven't already tried it, id strongly recommend running 'protocol override customize my settings' and alt-tabbing out to find the configuration dialog window.

I think you're misunderstanding the issue.
Terrible unrealistic example to give an idea...

You say 'deploy landing gear'. Response is 'landing gear down', but this is then misheard as 'landing gear' so the landing gear is retracted with a response 'landing gear retracted commander' which is misheard as 'deploy the fighter' so the fighter gets deployed. And so on.
Point is you said one command and half dozen or more other commands are randomly actioned as the mic picks up the crew responses and tries to reinterpret them into new commmands.

Ninja'd

 

[disorganise]

I have good headphones with excellent noise-cancelling mic but as I said I don't want to use headphones rather than my 5.1 surround system and sitting with the phones on my head / round my neck just to use the mic is a bit of a pain. It is not an issue at all when using headphones rather than the sound-system.

I wasn't clear. I meant that we should be able to have awesome noise cancelling since the computer knows the sound it is making, so it can completely cancel it out (to itself), leaving only your voice. IE, it is able to completely ignore the hcs voicepack output and the game soundtrack.

If headphones can do it in near real time by listening and responding, the PC by knowing the wave it is outputting ought be magnificent. But it doesn't
I did look for a tool that did it but didn't find any. Like you, I don't like playing with headphones on as a rule.

 

[Para Handy]

I have had a response on the HCS fora which is:

"Increase the confidence level or drop the command weight in your voice attack recognition settings. "

So once I learn how to do that then problem should be alleviated.

Thanks all.

I wasn't clear. I meant that we should be able to have awesome noise cancelling since the computer knows the sound it is making, so it can completely cancel it out (to itself), leaving only your voice. IE, it is able to completely ignore the hcs voicepack output and the game soundtrack.

If headphones can do it in near real time by listening and responding, the PC by knowing the wave it is outputting ought be magnificent. But it doesn't
I did look for a tool that did it but didn't find any. Like you, I don't like playing with headphones on as a rule.

I see what you mean - yes it would be handy. Thanks.

 

[Para Handy]

I think this is all I did, use at your own risk.

First I set up a Global Hotkey.

-----In voice attack options (wrench icon) select Hotkey tab.

----------Recognition Global hotkey, click on change box to open Global Hotkey window.

--------------- Check box for enable global hotkey.

--------------- Set a key or key combination for the global hotkey (I use LALT F12)

---------------Select Voice Attack listens while keys are down option.

----------------Click OK to close window.

----------Click OK to close option window.

Then edit the Singularity profile.

-----In the Voice Attack window, click on the pencil icon beside the profile name.

----------In the Edit a Profile widow, click on the check box beside the profile name.

---------------In the Profile Options window select the Profile General tab.

--------------------Check the box for “Override listening if my spoken command begins with:”

-------------------------Type the word or phrase you want to use (I use computer).

-------------------------Click Ok to close the window.

Apply change and restart voice attack.

Excuse me quoting all that again but I just wanted to put it out there for anyone else - this works perfectly.

Brilliant - My voice-packs crew are now working as intended and no spurious actions - thank you so much.

[up]

 

[Para Handy]

Just going to update this as I got fed up using the trigger word first...

I have a range of microphones (radio amateur and electronic engineer) so I spent a lot of time with different mics, mounts, baffles etc and was never happy with the results - still occasional voice-pack triggering each other (particularly that very loud Astra git). Now a contact does studio work so I organised the loan of some shotgun mics (including ones that you would think were made of platinum and Taaffeite by the cost) and no matter what I did the separation never totally overcame the voicepack output pickup, the same was true of some of their stage-use head-mounted wireless stuff.

In the end I just decided that I was going to have to put up with using the override / trigger word and went back to a normal mic. Somehow I forgot to switch off VA listening and found that amazingly, VA now was not responding to voice-attack output any more. Puzzled, I suddenly realised that I had connected a dynamic mic instead of one of my condenser ones. The dynamic mic also has a cardioid pattern, like my condenser ones, but of course dynamic mics are less sensitive than condenser ones.

TL / DR - Buy a dynamic vocal microphone and your troubles disappear (touch-wood).

(P.S. It needn't be an expensive one like a Sure SM58 - just a Behringer worked for me.)