Martin from TorontoFor blind folks, if you're getting annoyed with the word "boosting" in your feed and are using JAWS, consider adding the word to your speech dictionary and replacing it with an earcon (sound effect)... much faster browsing.
James Scholes@mcourcel Don't tell anyone, but I replaced it with the word "RT" in my speech dictionary.
@jscholes Lol! Another solid solution.
Pratik Patel
Drew Mochak@objectinspace I think there is an add-on, the name Unspoken is coming to mind. I've never used it, so not sure how customisable it is, or to what extent it might provide an equivalent experience. Personally, I'm terrible at processing speech and sounds together! @ppatel @mcourcel
@jscholes @objectinspace @ppatel @[email protected] this. Their's also one called Audio themes. https://github.com/SeanTolstoyevski/unspokenPy3/releases
@blindndangerous @objectinspace I think @modulux also recently posted a working build, although I'm not sure of their GitHub username. @ppatel
@jscholes @blindndangerous @objectinspace @ppatel Yes, my current build is at https://modulus.isonomia.net/doodles/unspoken.nvda-addon and the source, if you want to build it yourself or verify, is at https://github.com/bingen13/unspokenPy3
@modulux @jscholes @blindndangerous @ppatel There is also this https://github.com/Brynify/Unspoken/releases (thanks @fastfinge)... so you're telling me there are three different branches of UNspoken? That's amazing but also makes me wonder if the projects couldn't be murged at some point.
@objectinspace @jscholes @blindndangerous @ppatel @fastfinge I wouldn’t mind giving up to some other maintainer. I’m only keeping it up for my own survival so to speak, wasn’t aware there were other maintained forks.
Pratik Patel@modulux @objectinspace @jscholes @blindndangerous @ppatel I know of only two your version and the original. The original doesn't appear to have been updated since 0.4 release. unfortunately, I find it the add-on wildly inconsistent. My wish for a good solution remains.
@modulux @objectinspace @jscholes @blindndangerous @ppatel Oh! I had not seen that one until now. *rushes to get it.*
@modulux @objectinspace @jscholes @blindndangerous @ppatel Thank you for pointing it out. I do wonder why there are multiple versions wandering around.
@ppatel @objectinspace @jscholes @blindndangerous @ppatel You’re welcome, and no problem. Mine just exists because NVDA API changed and it needed to be changed to run at all. I assume someone else forked too and is doing a much better job than me at keeping the plug-in working. One random thing that I noticed is that changing the NVDA language (not the synth language) completely breaks the add-on because the role names change, for instance.
@modulux Almost as though relying on magic strings and monkey patching is a bad idea! Who would've thought it :)
@jscholes Not the best idea, but it has the virtue of working. I assume these roles are numbered somewhere, I just don’t know where or I’d use a number instead.
@modulux All roles and states are enumerated in the controlTypes module. Each NVDA object uses those enum values, and I feel like the original add-on author would've needed to use the same to look up a string representation. Unless they are being extracted from speech sequences. Shrug. You do what you have to do.
@jscholes Thanks, I should take a look and see if I can tighten the logic up a bit with less magic. I’m having to use NVDA in English even when not because it’s too insufferable to remake the sound theme or rename the sound files.