Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

NVDA has problem reading information with large dictionaries #1652

Closed
nvaccessAuto opened this issue Jul 6, 2011 · 14 comments
Closed

NVDA has problem reading information with large dictionaries #1652

nvaccessAuto opened this issue Jul 6, 2011 · 14 comments
Labels
Abandoned requested reports or updates are missing since more than 1 year, author or users are not available. bug component/speech

Comments

@nvaccessAuto
Copy link

Reported by drein on 2011-07-06 13:05
With quite big dictionary, NVDA is not able to make spelling, read date/time or copy to clipboard the t itle of a window.
It should very useful if some commands like insert+f12, insert+t have priority over the dictionaries, otherwise it is quite difficult to use the "double commands" insert+t of insert+f12.

@nvaccessAuto
Copy link
Author

Comment 1 by jteh on 2011-07-06 13:22
What do you mean by big dictionary? Are you referring to speech dictionaries? If so and your dictionary is large enough to break these commands, NVDA would be practically unusable, as we wait half a second for the next press of those commands. Are you saying that NVDA lags half a second before speaking any piece of text?

My preferred solution would be to disallow dictionaries of that size.

@nvaccessAuto
Copy link
Author

Comment 3 by drein on 2011-07-06 13:33
Hello,
yes, the italian dictionary of a sapi5 voice is about 375kb.
I'm not using that voice, but I gived it a test and NVDA takes about 1 second to speak after having issued a command.
you press insert+f12, wait about a second and then you have the time.
So do you think we should limit the length of a speech dictionary?
Thank you.

@nvaccessAuto
Copy link
Author

Comment 4 by jteh on 2011-07-06 13:38
Why such a large dictionary? Surely the synth doesn't mispronounce so much.

I guess limiting the size of speech dictionaries is the only option. I'm not sure how many entries to use as the maximum, though. That will take some experimentation.

@nvaccessAuto
Copy link
Author

Comment 5 by briang1 on 2011-07-06 14:14
Of course Sapi 5 is quite slow on lower spec machines to start with. it almost sounds as if someone is trying to make a voice for one language speak another one using a dictionary.
Even here I've put quite a bit into a sapi 5 dictionary and not slowed things much, so this one is really rather huge. More info as to why it is so big might be handy here.
Cannot say I've noticed any difference in normal dictionary sizes against speed.

@nvaccessAuto
Copy link
Author

Comment 6 by drein on 2011-07-06 15:58
Hello,
yes, Sapi5 Silvia hasn't a good engine, anyway what about a simple option/hotkey for the user that disable dictionary?
I'm thinking nvda+d or nvda+shift+d.
so when a person wants to read a long text, he enables his long dictionary, instead when he wants only move between menu for example, he disable it.
A part of this, I think that this feature could be good in general, not only for this particular circumstance, don't you think?

@nvaccessAuto
Copy link
Author

nvaccessAuto commented Jul 6, 2011

Attachment dict_toggle.py added by tspivey on 2011-07-06 16:18
Description:

Edit:
Added the attachment from the trac server, rename to dict_toggle.py
dict_toggle.py.txt

@nvaccessAuto
Copy link
Author

Comment 7 by tspivey on 2011-07-06 16:19
I've already done it as a global plugin, and attached it to this ticket.
NVDA+f11 turns the dictionaries on and off.

@nvaccessAuto
Copy link
Author

Comment 8 by drein on 2011-07-06 20:52
this is simply great. It works fine and I think it is a simply and easy solution. So, please don't limit dictionary size, since with this plugin user can decide what to do. Really thank you!
I only change hotkey from nvda+f11 to nvda+d, but this is subjective.

@nvaccessAuto
Copy link
Author

Comment 9 by m11chen on 2011-07-07 07:46
From my own experience, Traditional Chinese users used to use a 765kb speech dictionary for the speaking of character descriptions, but I have never noticed any lag in speech. Not sure why...

@bhavyashah
Copy link

@jcsteh @feerrenrut @tspivey Could the file attached in #1652 (comment) please be restored as well as the NeedsCodeReview label be added to this ticket?

@feerrenrut
Copy link
Contributor

Added the attachment to the comment, since quite some time has passed this will need to be re-tested, and a PR created.

@Adriani90
Copy link
Collaborator

@Drein are you still seeing this issue in NVDA 2019.2 RC2? @tspivey your global plugin is certainly a promising feature for some users. And if you could implement it with an unassigned key stroke that would be great. Are you planning to raise a PR for this any time in the future?

@Adriani90 Adriani90 added the Abandoned requested reports or updates are missing since more than 1 year, author or users are not available. label Dec 8, 2023
@Adriani90
Copy link
Collaborator

The performance of NVDA and of SAPI5 in in particular evolved gradually during last years, and we don't have any updates from the initial author on this issue after asking for it. I am closing this as abandoned. If you are still having this issue in NVDA 2023.3, please comment on it and we can reopen.

@CyrilleB79
Copy link
Collaborator

and we don't have any updates from the initial author on this issue after asking for it.

Cc @dreinn (with two n's), which is most likely the real initial reporter of this issue.

@dreinn:
As written by @Adriani90, if you still have the problem, please provide an update and we can reopen.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Abandoned requested reports or updates are missing since more than 1 year, author or users are not available. bug component/speech
Projects
None yet
Development

No branches or pull requests

5 participants