StenoTutor v0.1.0

[Emanuele Caruso]

I just pushed a new version of StenoTutor to github: https://github.com/caru/StenoTutor

Here the changelog.

Version 0.1.0

• Added on-screen keyboard with next chord highligthing. You can optionally disable on-screen keyboard, querty keys and next chord highlighting.
• Added optional text-to-speech word dictation, can be enabled in data/session.properties. Default: disabled.
• Dynamically show next line words as they are created to allow for a better typing flow at the end of lines
• Lot of refactored code, now it's much more readable.

Enjoy and let me know if you get into bugs or other issues!

[Mirabai Knight]

Emanuele, you're fantastic. Thank you so much for coding StenoTutor! The attention to accessibility is also very much appreciated, especially since one of the goals of the Plover Project is to reach out to aspiring blind and low vision stenographers (we've already got a few on board), who tend to have extremely good high speed audio processing skills from years of screen reader use; stenography can potentially be a very lucrative and rewarding career for some of these users. I'll try to post about StenoTutor 0.1.0 on the Plover Blog sometime today, if I get the chance.

[Emanuele Caruso]

Hi Mirabai,
thank you for featuring StenoTutor on Plover blog, I'm glad about this!

I agree about accessibility: it is not hard to implement and is key for a substantial portion of users. Now the text-to-speach support is very basic, but as soon as I implement a better command system I plan to add some features like commands to "repeat last word" or "spell next chord" etc. and also a way to have StenoTutor dictate x words ahead of the current target.
I also hope to get some feedback and ideas about the most needed TTS features from blind/low vision users.

I also see some benefits for non-native English users, since phonetics seems to be such an important aspect of stenography (correct me if I'm wrong): something that the QWERTY English typer doesn't have to consider in most cases.

[Mirabai Knight]

If I could submit a feature request, a corpus analyzer/lesson
generator would be amazing. You submit the full text of a Project
Gutenberg book (or other long text file) of your choice, and
StenoTutor could sort it into a histogram of words, with the most
frequent delegated to the first levels and the least frequent at the
higher levels. Words that aren't defined in your dictionary could just
be marked as *undefined*, giving you practice at making your own
definitions on the fly. You could practice either using the text file
as-is, so you could actually follow the story as you wrote, or you
could use the histogrammed version, practicing the most common words
first and the least common as you got faster and more fluent. It might
be fun to be able to switch back and forth, spending some time on the
intense, targeted word frequency drilling and then flipping over to
the more entertaining contextual sentence drilling, using the same
text file. Or if you're the sort of person who has your entire email
archive on your computer, you could paste that into one massive text
file, give it to StenoTutor to chew up, and then have a customized
corpus of your own most frequently used word, names, and jargon to
drill with!

[Stephen Laux]

Emanuele,

Thanks for putting this together. I am having some difficulty with the chord detection on my Sidewinder in the linux64 version of the app, as I find that StenoTutor will often type the underlying QWERTY letter in addition to the brief. For example, the chord -F outputs "u of". This contrasts with my text editor which correctly interprets the brief as "of". Any suggestions for troubleshooting?

Thanks,
Stephen

[Emanuele Caruso]

Hi Mirabai,
Everyone is encouraged to submit ideas and features request, and of course I am happy to receive one from you!

Your feature request is very interesting and I can see how it could be useful for a very wide audience (and it looks fun), I would like to implement a very basic functionality of it at first, when I find the time.
But let's suppose I can implement it all now: how would StenoTutor choose between multiple definitions of the same word in the dictionary? Perhaps show them all? Or do you have other ideas?

I hope you and the group will submit more and more feature requests. Probably I will not be able to fulfill them all, but it's nice to brainstorm, and anyway I hope that in the future as the Plover user base grows there will be more developers implementing new features, so it will not be bad to have a list of possible features.

[Emanuele Caruso]

Hi Stephen,
this is weird, because StenoTutor, like any other application, just listen to key releases generated by Plover for what concerns the filling of the input field...

Hesky, any idea why this is happening?

[Hesky Fisher]

On linux the typed character would show up first, then be backspaced and then plover's translation would be typed. Is StenoTutor correctly handling this backspacing?

--
You received this message because you are subscribed to the Google Groups "Plover" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ploversteno...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[Emanuele Caruso]

It should, each detected release of the backspace key deletes the last character of the input field... actually I am realizing now that for the backspace I should listen to keyPressed() rather than keyReleased(), but this shouldn't be an issue with Plover. Or maybe this could actually make a difference? Probably it is worth a try.
Stephen, does this happen often, not always?

I will try to do some testing on linux and see if I can replicate this.

Thank you, Emanuele.

[Stephen Laux]

The error occurs 99% of the time. It is very rare for the chord to output correctly. Once or twice I have gotten the program to correctly recognize the chord for "to".

[Emanuele Caruso]

I am sorry this happens on your system, I will try to figure this out.

Did you try to run the linux32 version of StenoTutor? Same outcome?

[Emanuele Caruso]

Hi Stephen,
I think I fixed this in the new version, please try and let me know how it goes: https://groups.google.com/forum/#!topic/ploversteno/xCOoNr_LZfM

[marn...@gmail.com]

Whenever I get to the end of the first line, Java crashes with an "out of memory" error in the animation thread. I've tried with both the Sun JDK and OpenJDK. What am I doing wrong?

T

[Emanuele Caruso]

This should be fixed in v0.1.4.