Replace most / all Google Voice functions with Tasker/Autovoice?

[Sunwind W]

Is there a simple way of doing this, or do you need to create a profile/event/task etc for every single thing?

Really, what I want is to be able to say:

"Computer"

turn on %ComputerAck variable, set to 1

<play sound beep>

*listen for command*

If recognized ->

<play sound beep 2>
*execute command*

If not recognized within 10 seconds:
<play sound beep 3>
Turn off variable %ComputerAck

The sound files being star trek computer effects. Yeah, I know, I know. :)

I've got it set up like that to do a simple thing such as open a program and then close it with another command, but the way I've done it seems very.. messy and cumbersome, there must be a simpler, cleaner way of doing things?

And I want to be able to set/delete alarms/timers with it like with Google Voice, but have no idea where to even begin with Tasker/Autovoice there.

Another thing that bothers me is that the recognition seems to be so slow. I want to be able to say "computer", and then have my phone recognize that and play the sound file almost immediately, but it seems to take on average 4-5 seconds, is there a way of speeding this up, or is this just how voice recognition is at this time?

I exported my profile, hopefully it gives a better idea of what I'm trying to do and how I've done it so far.
http://pastebin.com/0uBwycgs

Thanks in advance for any assistance.

[Brandon Horwath]

Is there a simple way of doing this, or do you need to create a profile/event/task etc for every single thing?

Not at all, this will use AutoVoice to input any set of vocalized process into Google now. This is simply an example of what I use. It will work for your needs, but as a part of a two part process I will explain.

Here is how I use AutoVoice to pass variables to Google Now, for everything after I say the first word of a command, "droid...". Somewhat similar to your desire to use the word, 'computer'.

Profile: # Droid REGEX (9)
State: AutoVoice Recognized [ Configuration:Command: "droid (regex)" ]
Enter: AV Droid Command (21)
A1: Send Intent [ Action:android.intent.action.WEB_SEARCH Cat:None Mime Type: Data: Extra:query:%avcommnofilter Extra: Package:com.google.android.googlequicksearchbox Class:com.google.android.googlequicksearchbox.SearchActivity Target:Activity ]

Remember action A1 carefully above!

Really, what I want is to be able to say:

"Computer"

turn on %ComputerAck variable, set to 1

<play sound beep>

*listen for command*

This is easily done. Create a aurovoice recognized profile for the word computer, exact. This will maybe say something like, "please state a command", (dying to know if you found a sound alike for Majel Barrett-Roddenberry) and set a command ID to something, the initiate a voice command prompt for the task.

If recognized ->
<play sound beep 2>
*execute command*

Simple, create a voice recognized profile for any command words with the last command ID set from the previous profile task set command ID. Then, this task action would be A1 exampled above, only using %avcomm instead of %avcommnofilter.

If not recognized within 10 seconds:
<play sound beep 3>
Turn off variable %ComputerAck

In AutoVoice, you can access the settings in the app and set a command ID reset time. Set this to 10 seconds.

The sound files being star trek computer effects. Yeah, I know, I know. :)

... sweet.

I've got it set up like that to do a simple thing such as open a program and then close it with another command, but the way I've done it seems very.. messy and cumbersome, there must be a simpler, cleaner way of doing things?

Try using AutoLauncher and other plugins for tasker to simplify this project.

And I want to be able to set/delete alarms/timers with it like with Google Voice, but have no idea where to even begin with Tasker/Autovoice there.

Hopefully, this gives you some places to start.

Another thing that bothers me is that the recognition seems to be so slow. I want to be able to say "computer", and then have my phone recognize that and play the sound file almost immediately, but it seems to take on average 4-5 seconds, is there a way of speeding this up, or is this just how voice recognition is at this time?

I think you can speed up the timeout for command recognition I n AutoVoice settings. And, I believe downloading the offline recognition does speed up the search query response time when using Google Now. But I guess, depending on your device/network you might not recognize a significant difference.

Hope this helps.