As many of you know, Vocola 2 is based on NatLink, which provides an interface to Dragon functionality via Python.  The current version of NatLink (4.X and earlier) uses Python 2; this is problematic, however because Python 2 is past its end-of-life.

 

Accordingly, Quintjin and crew are working on a new version of NatLink, 5.0+, that uses Python 3.

 

I've been testing Vocola against the new version, reporting bugs and adapting Vocola as needed.  As of today (8/7/2022), we are down to a few major blockers:

 

* NatLink 5+ is not correctly handling non-ASCII characters (SendDragonKeys, dictation with Vortex, execScript, etc.)

    * see https://github.com/dictation-toolbox/natlink/issues/113

* Vocola fails to compile and load changed Vocola command files at start of next utterance

    * this is both a failure to compile due to not getting begin call back and a separate failure to load changed grammars automatically by NatLink

 

Once those are fixed, most of the remaining work is just to package up Vocola as an installer or Python package -- the exact installer details are still in flux I believe.

 

Note that I will be maintaining support for the current version of NatLink for at least the next major version of Vocola.

Thank you for the update!

-------------------------

Win10/11/x64, AMD Ryzen 7 3700X/3950X, 64/128GB RAM, Dragon 15.3, SP 7 Standard, SpeechStart, Office 365, KB 2017, Dragon Capture, Samson Meteor USB Desk Mic, Amazon YUWAKAYI headset, Klim and JUKSTG earbuds with microphones, excellent Sareville Wireless Mono Headset, 3 BenQ 2560x1440 monitors, Microsoft Sculpt Keyboard and Logitech G502 awesome gaming mouse.

As an update: we (LexiconCode, Doug Ransom and I) are working on that Natlink release, with also Vocola2 and Unimacro. Dragonfly is ready for the new version, I think, Vocola almost, and Unimacro still needs quite some work to do.

As to the updating of changed vocola files: when you called for "edit commands" or "edit global commands", at each utterance check for changes is done.

When you open, edit and save a vcl file in another way, the changes will come through at mic toggle.

As an update: we (LexiconCode, Doug Ransom and I) are working on that Natlink release, with also Vocola2 and Unimacro. Dragonfly is ready for the new version, I think, Vocola almost, and Unimacro still needs quite some work to do.

As to the updating of changed vocola files: when you called for "edit commands" or "edit global commands", at each utterance check for changes is done.

When you open, edit and save a vcl file in another way, the changes will come through at mic toggle.

Thanks for the update guys & much appreciate your work on this !

Quintijn, Mark, any update on the current status? It looks like progress has been made based off of reading the gitter, but I can't quite tell of it's in a release state?

Update: Doug has fixed the NatLink 5.3 bug where it was incorrectly handling non-ASCII characters. Unfortunately, in the process of testing it I discovered that my SendInput implementation also did not correctly handle these characters under Python 3 :-(

A lot of installer changes are being made for a better experience (e.g., Vocola and the like will come as python pip packages); rather than wait for all that work to settle down I'm going to make Vocola 2.8.8 a separate download from the installer. Full integration probably won't come until 2.8.9.

UPDATE: I fixed my SendInput implementation today.

Now I just need to figure out how to get commands be compiled and loaded when a Vocola file is changed...