Text to speech for readback about atc HOT 10 CLOSED

@zlsa what icon pack are you using for the controls? I have been looking into implementing ionicons, but I didn't want to change them over if you were already using an icon pack.

https://github.com/driftyco/ionicons

from atc.

The icons were all handmade in Inkscape. If you want any additional icon, I can make it.

from atc.

I had a play and added speech synthesis for readbacks using the W3C web speech API. There's no on/off switch and on my machine (Chromium 39 on Linux) the speech gets stuck after every readout so I have to call speech.cancel() manually each time but it does read things out. Also I needed to start Chromium with the --enable-speech-dispatcher parameter for it to work.

https://github.com/daviessm/atc/commit/116836f7fa70c09352b563808e1b9c3e0c2791ff

from atc.

@daviessm hahahaha! I just cloned your repo to test and I gotta say it's pretty cool! I'd be amazing if the voice could be a little more "natural".

Still pretty cool.

I don't think read-back should be on by default but if we could add a button to enable/toggle this and maybe work on the voice it'd be a great addition!

from atc.

@Diastro The speech is done through the browser which calls OS libraries to do the actual synthesis. How it sounds will depend entirely on that. I've just tried it on my Android phone (5.0.1) and it sounds pretty natural.

Agree about the on-by-default but I suck at UI design so I'll leave that up to someone else. It would be easy enough to add another variable speechOn or something to speech.js and toggle it.

I would also like to have it pick random voices for each plane. So some are male and some are female. And have the tower controller as yet another voice.

from atc.

@daviessm Gotcha. Yeah regarding UI @zlsa is probably the best person to help for that last part.

from atc.

Can we close this?

from atc.

@smp4488 Sorry to have kind of jumped past you on this - does it do what you were thinking of?

from atc.

@daviessm it's all good. Holidays have me all tied up. I haven't had a chance to look at your changes yet, but you can see what I have added here https://github.com/smp4488/atc. I created a voice / voices model wanting to offer the user a choice of the available system voices. I have most of it hashed out, but the ui still needs a little love.

from atc.

@smp4488 The additions I made weren't configurable at all - it just adds a call to the speech synthesis API. @zlsa added an on/off toggle to that. I think it'd be worthwhile to have a UI for changing other settings too.

from atc.