In order for Jarvis to understand us, we need a system that takes voice commands. Google Now is an excellent start for this. But as said, we need to be able to customise it. We need to add our own commands and make this as simple as possible. Therefore, say hello to my little friend “Autovoice”.
Autovoice is a plugin for Tasker which is built on Google Now. As you have probably already used Google Now and have seen what it does. Imagine that you now can add any command to it and it will execute what you want it to do.
The price for Autovoice is around $3 and that’s in my opinion for free. I have tried to find the limits of this plugin but I haven’t come across any. The best $3 I have ever spent.
I mainly use Autovoice Recognized as I don’t really see the upside of the other possibilities like Natural Language. If you can convince me otherwise, please do!
I’m not going to explain how you have to set up it the first time because Lifehacker has a much better post on that. So, you can always check there.
But I will list here a couple of features that you really have to know in order to make your life a lot easier. Create a new Profile => Event => Plugin => Autovoice => Recognized. Then I always choose The Hard Way as I’m used to it from the past and it gives you also some very interesting extra features.
2 features are really key for me:
Use Regex: A regular expression is a pattern that the regular expression engine attempts to match in input text. Like you see in the image below. I said the word “test” and Autovoice is going to look at the pattern and if it matches an Autovoice Event Recognized profile, it will launch the task associated with it. So if you feel that Jarvis doesn’t understand you well enough, you can always switch this on for a better result.
But my favourite feature is actually Contains All. The explanation is actually quite clear in Autovoice itself: Check to make the command match if all words in the command filter are contained in the spoken command, no matter which order they appear in. Why do I like this that much? I have already a lot of Autovoice Profiles. So many actually that I even forget from time to time how the command actually went. In order to bypass this, I just add 2 or 3 words which I know I will always say and then the command will be launched. A good example of this feature is my post on Sports Headline of Today, I posted the video at the top of this post. Here, I just add the following 3 words to Autovoice: “sports headline today”. So every time I say any sentence containing those 3 words that task will be triggered. That’s pretty neat, isn’t it?
Autovoice is in my opinion amazing but there’s one feature that I would like to see even more developed: Autovoice Continous. If you switch this on, Autovoice is always listening but there a couple of hiccups that come with it:
As Google Now has that beep, with Continous, Google Now is then always on and that annoying beep comes every 5 seconds. So you have to mute your phone/tablet. You can do this with a command of course. But it is not that user-friendly.
It would be great that my Android is always listening so I don’t have to press the button every time. You can bypass this by saying “Ok Google” but this is not the same. I like to say “Jarvis” and then Google Now should pop up. There’s an app called Open Mic+ for Google Now where you can adapt this but this app seems to crash every time on my device so unfortunately no luck there.
To conclude, Autovoice is an amazing app and I’m expecting in the future a lot more out of it. There’s only one hiccup at this moment and this is Continous but this is not crucial to my setup.
How are you using Autovoice? Please let us know in the comment section!