Control Tasker With Voice? (specific context within)

[Brendan LaFord]

Hi there. I've been using Tasker for some time. I'm no expert but it
seems every day I find something new to do with it. I have a very
specific task I'm trying to automate currently.

I love my new car, but. . . the Bluetooth system is limited and does
not allow for in dash reading of SMS. In addition if I press the
phone button on the wheel, it merely allows me to call someone and not
send a text via voice. Annoying, considering I don't like to talk on
the phone. . . ever.

So, I've created a profile where if I am connected to my car and I
receive a text message the phone will read aloud the name of sender
and the message. I've also copied that dandy one that sends an
automated message as to how fast I'm going.

Now this is fantastic, but I can't help but think that there may be
certain situations where I most certainly DO NOT want the message read
aloud. Since the bluetooth system cannot tell my phone that the
passenger airbag is activated and thus someone is in the car with me
(Or can it?), I would like to have the choice of reading the message
aloud. And since I have thus far managed to avoid touching the phone
or reading while driving, I would like to give this command via voice.

Is this possible? I have found the "Get Voice" option, but how do I
link my speech to a variable value so that Tasker will know whether or
not to read it aloud?

Ideally what I would like to happen is this:

1. I'm driving and receive a text message.

2. Tasker says "New message. Read aloud?" (Let's face it. Hearing
who the sender is could be just as bad as
hearing the actual message. :p)

3. At which point I say yes/no or, even better, I give my best Jean
Luc Picard "Make it So!" (or something equally nerdy) and Tasker
recognizes this as a yes.

4. Tasker then says: "Message from %SMSRN: %SMSRB" if indeed I gave
the correct command to do so.

5. Tasker then says: "Reply?" I say "Yes" and then it listens for a
predetermined time period after which it reads to me what it heard and
asks if I'd like it sent. Brilliant!

I was thinking too, and I'm saying more than I should, that it would
be really cool if when I said "Yes" to the initial read aloud question
that it immediately respond something like "Error, please try later."
Or, better still, it would say "Message from Mum: Thank you so much
for the lovely flowers." Seriously I'm rambling and I'm not that
awful.

Thank you for you assistance. Any ideas are welcome.

[Brendan LaFord]

I should mention I have the latest version of tasker, a verizon
wireless droid x, and I'm running android 2.3. And If anyone is
familiar with the Bluetooth system, the car is a US 2011 VW GTI.
Thanks again.

[nanobrain]

The result of your Get Voice action is stored in the %VOICE variable (for more info check the help for the Get Voice action), so your profile would look something like this (disclaimer: I haven't really tried it out)

Profile: Incoming text

Task: 

Say "New Message. Read aloud?"

Get Voice (make the timeout like 5 seconds)

IF %VOICE ~ *yes* (or *make it so*) :)

Say "Message from %SMSRN: %SMSRB"

Say "Reply?"

Get Voice

IF %VOICE ~ *yes*

Get Voice

%SMS_MSG=%VOICE

Say %SMS_MSG

Say "Send?"

Get Voice

Send SMS with SMS_MSG IF %VOICE ~ *yes*

ENDIF

ENDIF

If you want it to react to a "yes" only leave out the asterisk in the matching field. You probably need to put in some logic to jump back if you are not happy with the dictated SMS and a way to abort the whole thing.

Good luck

 

 

 

[Brendan LaFord]

Fantastic, I can't wait to try this.

Do you happen to know, is there a way for tasker to distinguish
between merely being connected to my car via bluetooth and streaming
audio via bluetooth? With my system, for instance, if I am listening
to a CD I will still get phone calls over my car's stereo but will not
get other notifications from phone (navigation, sms, alerts, etc)
unless I am specifically using bluetooth as my audio source.

I would love to use the above described task when I am streaming audio
from my phone. When I am not I would like to use the automated
message that I am driving at x mph etc.

[Brendan LaFord]

In actuality this turned out a rather slow and painful process. The
"get voice" takes a few seconds just to initialize. So not worth it.
But I'm thinking a call back feature would be fine too.

Is there a way to have Tasker auto dial the number of the last SMS
sender? I am thus far having no luck as it dials what appears to the
contact name spelled out in touch tone keypad numbers.

[nanobrain]

Use %SMSRF instead of %SMSRN

[V Oz]

Come on! You should do next:

1) get a picture with front camera,

2) recognize are there passengers in the car,

3) if there are, recognize are they trusted to hear Your sms or not.

4) depends on p. 3 read or put it into FIFO buffer to read later.

hmm. it's a kind of joke.

Good luck, 007 ;)

[Brendan LaFord]

I like your style. Not a bad idea at all.

[Kevin Taylor]

Step 4 should be:
If passengers are untrusted, translate sms to Elvish before reading.
(unless passengers are seen to be elves)

Kevin