nvaccess
Provision of indication options for reporting spelling errors.
Reported by dhankuta on 2014-06-29 06:46 Some times the reporting of spelling errors disturbs specially in paragraphs with many proper nouns like name of person/places. If reporting option set in off mode, it may be useful for another paragraph. A repeat of on/off of reporting mode is not practical too. It is often irritating to listen the phrases 'spelling error/out of spelling errors' but an indication of wrong spelt words is equally fruitful. If options of saying the phrases as it does now, a choice of an indication by a beep and no reporting (similar of reporting the uppercase character!) is adopted; situational setting becomes easier. I have made the changes as above of setting the reporting of spelling errors from Boolean to combo box selection and I am feeling easy on handling different situations. Is it adoptable in nvda main stream? See also #4877 See also #4633
Comment 1 by jteh on 2014-06-29 10:04 It's not currently possible to have beeps or sounds at specific points during normal speech. This means that what you're suggesting won't work, for example, if there is a spelling error in the middle of a line. As you point out, this does work for capitals, but spelling is handled as a very special case and can only read individual characters.
I have some ideas about refactoring our speech code to allow this kind of feature, but that is a major chunk of work and I don't have any ETA as to when it might be started.
Comment 2 by dhankuta on 2014-06-30 05:48 Hi, Thanks for the quick reply. Perhaps my writing may have some slightly confusing statements. I understand the limitation and unusual beep while a normal reading. An insertion of a beep inside a text composed by joining many conditional parameters of a field is not possible. What I mean was not while reading a line. On reading a line it will be either on or off mode. But the just a beep is while navigating a word or character which I think is possible now. I think it is useful if the getFormatFieldSpeech function be redefined as many unconditional get functions like get style, get pagenumber, get color, get font, get spellingerrors etc. Based on the config values, we can join the returns of these functions as per need of a joint text. With these changes; we can have many features like just knowing a single parameters like only the color or style, or a single line etc... I have done it in some parameter and using it. My suggestion is to extend it.
Him Prasad Gautam.
Comment 3 by jteh on 2014-12-02 04:06 I think it's confusing to have beep functionality which only works when reading by word but it reverts to speaking when reading by line or larger. We should wait until speech can support this properly.
Will there be any movement on this?
I have a blind student that complained about this. He would like an option to here a buzz or beep when reading lines with spelling errors. He is a Brazilian Portugueses user and it's quite annoying/inefficient hearing "erro ortográfico" while reading a document. He think a noise would be a better contrast with the reading of the text.
Thanks for all your great work! Brazil loves NVDA!
Yes what we need is an extra setting. At the moment we get a sound when you type a wrong spelling but not when you are cursoring over it. Brian
[email protected] Sent via blueyonder. Please address personal E-mail to:- [email protected], putting 'Brian Gaff' in the display name field. Newsgroup monitored: alt.comp.blind-users
I think I'd like to see the noise it makes when you are typing an error carried over to when you are reading it with cursors etc, but both it and voice taken off for say all really. This may need multiple choice combo boxes here, but I think it would be better than the current mixed up approach. Brian
[email protected] Sent via blueyonder. Please address personal E-mail to:- [email protected], putting 'Brian Gaff' in the display name field. Newsgroup monitored: alt.comp.blind-users
cc: @feerrenrut can you remove the blocked term here? Now that speech refactor has been introduced, this issue is not blocked anymore.
There is no blocked label. I updated the description to remove the word blocked.
This issue is open for more than 10 years and having it fixed would really be useful.
PR #10474 was opened more than 5 years ago as a solution for this issue; thanks @jcsteh for this first work. Unfortunately it has been closed as abandoned.
In order to have this issue closed with the minimal possible discussion, I'd like to take over #10474 and add the needed configurability. For this, I plan to replace the "Spelling errors" checkbox in document formatting settings by a combo-box, as for line indent setting, offering the following options:
- disabled
- speech
- sound
- (optionally) both speech and sound
@seanbudd would you accept (triage) this solution?
@jcsteh or @LeonarddeR are there other things to update in the abandoned PR? I'd like to focus on the minimal requirement to avoid delaying this 10 more years (so no full audio theme refactor suggestion please!)
@CyrilleB79 happy to accept the solution proposed in #17997. however we don't feel comfortable triaging this issue in it's current state, as the ticket itself doesn't contain up to date information. Feel free to write up a new summary of the feature proposal and we can update the issue description or open a new issue and triage it.
