Toolbar and Commands about speech-to-text-in-tw5 HOT 18 CLOSED

from speech-to-text-in-tw5.

Hi @flancast90 , what we could do is let users create tiddlers tagged with $:/tags/VoiceCommand, let them add a field voice-command with the corresponding text that should invoke the action and in the text field they can write the action which should be invoked

from speech-to-text-in-tw5.

That's a good idea! What if we just add some kind of $:/config/VoiceCommands tiddler, with some input fields and a button to add a new field onclick. The input fields would define the input (words), as well as the code/action the words will invoke.

At runtime, the program can listen for 'Ok Wiki', and then detect if what comes after that is defined in one of the fields. If it is, it can run the code in the input next to it (store the number of the input with data-id or similar, possibly). If you can have the program create this tiddler (if it doesn't exist already in the user's wiki), and give me some way to grab the inputs (by id or some other method), I can do the rest!

from speech-to-text-in-tw5.

I'm one step ahead @flancast90 - see #7

from speech-to-text-in-tw5.

Pasta Macaroni Sidebar Question: just tested this, and nothing seems to happen. Is this intended? What is supposed to happen?

from speech-to-text-in-tw5.

😁 when nothing happens then it did not recognize one of the words correctly
the sidebar should toggle open/closed

from speech-to-text-in-tw5.

Did you get a resulting transcript tiddler @flancast90 ?

from speech-to-text-in-tw5.

I just made "Pasta" an alias for "Ok" and "Macaroni" an alias for "Wiki" (not within the JS code, user-side) ... and a user-defined action that is triggered by "Sidebar"

from speech-to-text-in-tw5.

from speech-to-text-in-tw5.

That's strange @flancast90 , on my side it's working fine. Does "Ok Wiki Sidebar" work for you?

from speech-to-text-in-tw5.

We'll probably re-add some console logs and see their outputs

from speech-to-text-in-tw5.

Well I can't even test it now :) The thing either thinks: wiki is Nikki (got wiki as Google once, too), or Sidebar is: divebar, Side Bar, etc.

from speech-to-text-in-tw5.

You can use the tiddler $:/config/speech-to-text/keywords and add "Nikki" to the field wiki-keywords
And you can use the tiddler $:/VoiceCommands/toggle-sidebar and add "divebar" and "[[Side Bar]]" to its voice-commands field

from speech-to-text-in-tw5.

OK Wiki Sidebar works, just had to refresh and move my phone away from my laptop Mic. I'd assume the closeness of the devices was causing the Mic signal and my voice to be distorted to the program. Probably worth mentioning in the documentation, too.

from speech-to-text-in-tw5.

Worth noting, too, that side bar is recognized as salad bar 75% of the time. Is that easy enough to add to the JS right now, as with Wiki? Or does it need new code for that?

from speech-to-text-in-tw5.

Worth noting, too, that side bar is recognized as salad bar 75% of the time. Is that easy enough to add to the JS right now, as with Wiki? Or does it need new code for that?

Yes 😁 that happened to me, too

You'll have to understand the user-voice-command mechanism which is done in WikiText only

from speech-to-text-in-tw5.

Sidebar is a custom user command
It's in a tiddler tagged $:/tags/VoiceCommand
The JS side filters all tiddlers tagged so and stores whatever is in their voice-commands field in an array. What's in the text field is stored in another array (the actions that should get triggered when one of the voice commands from the voice-commands field is detected)

from speech-to-text-in-tw5.

I'll close this issue and create a new one for the text-editor toolbar button only

from speech-to-text-in-tw5.