Capture open-ended response text as variable
under review
R
Richard
I love everything that has been added so far, really remarkable! However, it would be a game changer if for the Question action with regard to Flows if it was possible to ask an open-ended question and to capture whatever text was spoken back to the device and assign it to a variable. This would be a game changer, with the captured text then able to be sent to another service via webhook for example.
N
Nathan
I also believe this would be a game changer.
I have just migrated to VoiceMonkey from IFTTT, and was able to replace all but one IFTTT applet. The ability to capture and use Alexa speech-to-text within a VoiceMonkey flow would not only allow 100% IFTTT applet replacement (in my case), but allow seemingly limitless future development.
My current use case:
I have a small app running on my home network that interfaces with the Remember the Milk API. In order to add tasks, I need to be able to capture open-ended response text. The VoiceMonkey flow would work something like this:
Alexa Routine: "Alexa, add task" // ask vm to start flow n
Start Flow: Prompt: What task do you want to add? // "get oil change"
Assign Variable: task = "get oil change"
Web Request: to app, with task as POST or url-encoded GET param
The app then adds the task to Remember the Milk using their API.
I know there's an RTM Alexa skill, but for some reason "Alexa, tell Remember the Milk to remind me to get oil changed" is just too much of a mouthful for me to use practically.
C
Christian
marked this post as
under review
Richard Saurabh Tyagi - thanks for the feedback. Great news is that this is already on my list. Now I know there's demand, I'll make it more of a priority.
Rhys - If I understand correctly, I think what you're looking for already exists. When you create a Flow, select the "Question Prompt" action. Select the "First Name" answer type and add the possible name answers e.g. "Rhys", "Joe" etc
Then click on "Additional Settings" and under "Assign response to variable" to you can select a Variable you've previously created to assign the name answer to. You can then use this variable in other places.
Let me know if I've misunderstood.,
R
Richard
Christian: Hi Christian, thank you very much! I'm sorry it's taken me a moment to reply here. It seems what you've suggested (using the "First Name" option under the "Question Prompt") is partially what I am trying to accomplish however I would like to be able to capture more than just one single word. I'd really like to be able to capture simple phrases or a couple of sentences or two at a time (I realize there may be limitations from what an Alexa skill will allow with regard to capturing spoken audio).
R
Rhys
Saurabh Tyagi I was going to suggest this, but came across OP. Again, kudos for this system it's made using Alexa for my IoT actually bearable.
I have a security system set up using a clusterfuck of routines, alexas, flows, cameras, etc. but something I want is user recognition. If I could set a string variable using voice for a name as a response, would greatly improve my flow.
Other ideas for how to use this would be:
-Calling and texting using Alexa (properly). Recipient name and message etc. could be variables.
-User defined preferences to belt-feed to Alexa
I don't have any more examples to mind, but this is two people who have asked for this now. Got to mean it'll be used somewhere ¯\_(ツ)_/¯
S
Saurabh Tyagi
That is a great idea, Richard. Would you mind giving some examples of where will you use this?
N
Nathan
Saurabh Tyagi , Christian
Sorry I meant to reply here, but accidentally replied to the OP. I included an example use case in that previous post. But really, once you have those text strings, you can build or interface w/ essentially any API using VoiceMonkey and your voice as input. That would be pretty powerful.