Using Voice Input TiddlyWiki

[steved]

new user here - thoughts on how to use voice annotation for TW input? Currently, on my Android phone, I use google assistant and can simply speeak to make remionders, notes and lists which go neatly into Google Keep - they Copy Paste into my TW setup - so Keep is part of my setup. Is there more direct way people have done voice input. EG A tool that reads Keep as input and adds it formatted into  TW. 

[Mat]

simply speeak to make remionders,

Really, I'd never guess ;-) ;-)

Sorry, couldn't resist. Anyway, I'd say it is a browser matter so just do a search for your browser (example). 

<:-)

 

[@TiddlyTweeter]

Edge on Windows 10 has pretty good voice control, including voice typing, once you have trained it. I think that is because its more than a standard browser, being close to its OS.

Or you just interested in Android solutions?

Best wishes

Josiah

[TonyM]

Josiah,

Given your experience with voice dictation on edge would you be able to give us some high level instructions on how to do this?

I can then experiment and try and find ways to use it to integrate with tiddlywiki. 

Regards

Tony

[@TiddlyTweeter]

TonyM wrote:

... would you be able to give us some high level instructions on how to do this?

I'll write a separate thread on it because I think that steved was most interested in Android, not Windows.

Josiah

[@TiddlyTweeter]

Mat wrote:

simply speeak to make remionders,

 

I'd say it is a browser matter so just do a search for your browser (example). 

Still some merit I think in explaining it a bit because voice recognition has improved so much in recent years--but unless you used it it can remain the obscure option it isn't any longer :-).

Its interesting, for instance, Chrome, when using Google Docs, that has a specific mode for "voice typing", has pretty good accuracy for long dictation.

I like voice typing, especially for longer diatribes, as I'm a  "one-finger-typist". So, with training, its quicker than my lack of dexterity.

Side thoughts

Josiah

[Watt]

I'm dictating this into the body of a Tiddler using voice typing on Android and as you can see accuracy is nearly 100% . On my Android phone, in English, input to tiddlers with voice typing is easy and amazingly accurate. The problem is with navigation, you can't open a Tiddler using voice and moving around the Edit interface isn't possible.
If it were possible to map keyboard shortcuts to custom action buttons, and voice commands to keyboard shortcuts, that might work but some thumbs will always be needed I think. It would be another big leap for TW on Android when combined with Quinoid though.
Steved may prefer a method of importing from Keep. Maybe somebody on the forum has experience of doing that. What formats does Keep export to Steved?

[@TiddlyTweeter]

Right. Using a vocal navigation system alongside a dictation system is still problematic.

I'm not so clear if this is a system issue or a local switch. Thinking about what is involved it seems something about levels of recognition. It may be a standards issue? Do they exist?

Josiah

[steved]

Folks have mentioned basic voice dictation (speech to text) and systems that will do speech to text accurately in any open text area including in an open tiddler - and yes that all works, in chrome, in windows (with IFTTT)  and on android and iphone. But I was trying to hit the next level (which I have now in google assistant) - where I can be in a car or walking and simply say "hey google" ( or Siri, Alexa) and get high level control and knowledge put correctly into a note.   So Keep  (google changed it now it is google lists ) is nice on google with an assistant because I can say at any time to my phone and now pc "take a note, name it and add to it and even later have it read it back, or add a reminder ot appointment to my calendar.  So as I move to TW as my main personal knowledge system, I was hoping for that level of significant ease of voice interaction - or trick to achieve. One trick is to still use google assistant to make a list or now - but somehow write a 2nd tool to move it to a new tiddler ( again IFTTT or some background process might do this), again I am currently just being organized about what I call my voice notes so it is very easy to move them into my TW. So what I am doing now,  I say take a note to assistant and call in 't todo' and use my voice to put in todo I think of , or make a note of a new idea that I call t <idea name> and chat on about that idea. Then typically on my laptop/desktop. I have keep or lists open and I simply look for thing s with "t" in the beginning and move them into tw tiddlers. 

As was asked about Keep format - you can "copy to Goolde Drive", which is great cause it gives you a url and plan text - so possible to write something to copy in to a tiddler. 

[Mark S.]

Don't you need an always-on data connection for the voice/dictation? Just wondering.

Do your voice commands let you share your "keep notes" with another app? If so, you could tell G/Keep to share with an android-based TW app.

[steved]

I am using g/Keep now - via google asstiant, I anyway reminder myself of things or make lists with pure voice which puts in G/Keep and G/notes but doing copy and past to TW. if I continue to do that I guess I can try and write a better way to auto move over.