Tasker + VoxCommando=Practical Artificial Intelligence (tutorial)

[rebelmaveric19]

My version of getting tasker on android in conjuction with autovoice and autoremote to use as a speaker and mic for VC. I am not very good at writing tutorials and probably not really good at making video tutorials......I hopw you all can make something of this......

https://www.youtube.com/watch?v=cNvryvgGeEI&feature=youtu.be

Some thoughts after using it a few days:
1. Sometimes dosent seem like I get a response back......trying to figure out why
2. VC webserver has no kind of authentication (if using outside of lan it is a problem).............maybe james can help or tell me some simple way to provide authentication
3. My brain is going a million mph thinking of possibilities and cant sleep

[jitterjames]

This video has been removed because it is too long. 

[rebelmaveric19]

Had to verify my account.? Why didn't it ask BEFORE I uploaded the video. Don't understand some of google's way.......... video works now.

[jitterjames]

It works for me now.

[rebelmaveric19]

Does that video make any sense to y'all. I know you (james and nime5ter) probably understand it already.

[keithj69]

Does that video make any sense to y'all. I know you probably understand it already.

As an extremely new user to vc, I thank you for the video.  It answered 1) can i use tasker with vc somehow?  2)  can i do it remotely with tasker? 3) what can i do with eventghost? (lack of looking on my part). 4) can autovoice/vc understand one if they have a dip in? 

as a side comment, you could always set up to icon to turn on autovoice and one to tun it off.  I usually have autovoice on at night to send commands to my vera, just in case.

In all seriousness, thanks for the video.

[rebelmaveric19]

Thank you keith for the nice words....

By the way I do have a quick tile shortcut for av continuous and I do the same thing at night, have a profile at night called alarm clock and turns on av continuous so I can tell it to shut up in the morning.

[jitterjames]

Hey looks very promising.  I've been up to my ears in code yesterday and today so I have not been able to watch the entire video from end to end but I did jump through most of it and it looks interesting.

I'm trying to figure out if there is a way to get the second "alternate" from autovoice.  I notice that a lot of the time the recognition fails because it is using Google's recognition which does not know what the correct phrases are, so for example, when you say "Jarvis no" it just makes a wild guess that you said "Jarvis know" which VC would never do.  This is why VoxWav and VoxCommando generally work a lot better, but since you are trying to do this remotely I can see that it is worth trying to make it work an alternate way.  It may be possible to send VoxCommando multiple recognition phrases and it will go through them until it finds a valid phrase.  Most likely Google is returning both "Jarvis know" and "Jarvis no" as possible phrases, but guessing wrong about which one is the most likely match. I am going to try to take a look at autovoice to see if there is another variable that can be used.  I notice that avcomm is the "first match" or something which would imply that there are more possible matches.  Do you know if these are available to Tasker as variables?

When I first started making VoxWav I created an app called VoxCommandroid which used the Google recognition engine.  It employed this technique of sending multiple matches to VoxCommando.  I ultimately did not continue development though because I was not impressed by the poor results I was getting compared to a microphone with the VC engine so I switched to steaming audio which was much more difficult to code but ultimately yielded the results I was looking for.

For now your setup looks like a good choice for anyone that wants to talk to VC from a remote location, but it would be nice if we could increase the accuracy a bit.

[rebelmaveric19]

I am going to try to take a look at autovoice to see if there is another variable that can be used.  I notice that avcomm is the "first match" or something which would imply that there are more possible matches.  Do you know if these are available to Tasker as variables?

Yes it is %avcomms()

Also is there anyway to setup some sort of authentication on the webserver? Or should I worry about it?

[jitterjames]

There is no way to do this in VC at the moment.  I don't know if you should worry or not...  I guess it depends on what kind of commands you have in place but you probably have reasonable security through obscurity at the moment.

I don't know if there is some other way to secure this.  I guess you could set up a vpn or something but I don't know much about that stuff.

[jitterjames]

I have it working but after I send the http get to VC from tasker I get an error:

java.io.EOFException

Do you also get that error?  Is there a way to turn off the error popups, or better yet do you know how to fix it?  I can't find anything online about this error for tasker http get actions.

[rebelmaveric19]

Yea just turn on "continue task after error" in the http get command and it quits popping up that error.

[jitterjames]

Thanks.  I was not seeing that option until I turned off beginner mode.

Here is my task which sends multiple phrases to VC. It still isn't perfect, but you are more likely to get a hit.

EDIT added 21-11-2015 (nime5ter):
For an official VC tutorial on using Tasker w/ VC, please see: http://voxcommando.com/mediawiki/index.php?title=Using_Tasker_with_VoxCommando

[rebelmaveric19]

I think that error popups because VC simple server has no info on it. Somebody that knows java might be able to help more.....

[PointCloud]

Thanks for the video demo of this.  I like how you explained it can be used over the internet to gain controlled function and not just lan.   I also see a lot of potential with Tasker and Vox working together to make practical and useful tools.  I am going to sub to your channel if you decide to document more of your progress.  It is very useful information. 

cheers!