1. nvda@nvda.groups.io
   Date   
Re: nvda and app windows not closing

Joseph Lee
 

Hi,

That’s because universal apps are not really “closed” when you press Alt+F4 – they are suspended, hence these windows will show up via object navigation.

Cheers,

Joseph

 

From: nvda@nvda.groups.io <nvda@nvda.groups.io> On Behalf Of Josh Kennedy
Sent: Tuesday, July 17, 2018 11:29 AM
To: nvda@groups.io
Subject: [nvda] nvda and app windows not closing

 

HI,

If I run some apps such as settings, alarms and clock, calculator, store, and so on, and then I close everything out with alt f4… If I use object nav on the desktop set to simple review, and I go up to the list view and go across, it still shows settings window, calculator window, store window, and so on. And I have to go into those objects and find the close buttons and hit capslock enter to close those windows. Why is it that even if I close out apps, some sort of windows stay open on the desktop? Are they notification windows? They can’t be notification windows, because I went into notifications and cleared them all out and away. So I have no notifications. So what are they and how do I keep or prevent them from appearing? I also have an object which is after the desktop list and it just says “window”. When I OCR it it shows some desktop items but not all. I wish we had an advanced object manager that let us reclass, relabel, and even hide various objects, and also do certain things when objects appear or disappear.

 

Josh

 

 

Sent from Mail for Windows 10

 
nvda and app windows not closing

Josh Kennedy
 

HI,

If I run some apps such as settings, alarms and clock, calculator, store, and so on, and then I close everything out with alt f4… If I use object nav on the desktop set to simple review, and I go up to the list view and go across, it still shows settings window, calculator window, store window, and so on. And I have to go into those objects and find the close buttons and hit capslock enter to close those windows. Why is it that even if I close out apps, some sort of windows stay open on the desktop? Are they notification windows? They can’t be notification windows, because I went into notifications and cleared them all out and away. So I have no notifications. So what are they and how do I keep or prevent them from appearing? I also have an object which is after the desktop list and it just says “window”. When I OCR it it shows some desktop items but not all. I wish we had an advanced object manager that let us reclass, relabel, and even hide various objects, and also do certain things when objects appear or disappear.

 

Josh

 

 

Sent from Mail for Windows 10

 
Re: Tips for speed reading /listening with screen readers

Jim Homme
 

Hi George,

I would love to try out this voice. Can you please give instructions about how to install it? And can you point to information about how to change and test voices Espeak uses?

 

Thanks.

 

Jim

 

 

 

==========

Jim Homme

Product Manager

Digital Accessibility

Bender Consulting Services

412-787-8567

https://www.benderconsult.com/our%20services/hightest-accessible-technology-solutions

People with disabilities, access job openings at https://www.benderconsult.com/careers/job-openings   

 

 

From: nvda@nvda.groups.io [mailto:nvda@nvda.groups.io] On Behalf Of George McCoy
Sent: Tuesday, July 17, 2018 11:57 AM
To: nvda@nvda.groups.io; acsociopath@...
Subject: Re: [nvda] Tips for speed reading /listening with screen readers

 

Greetings,

 

Unfortunately, none of the espeak variants that come with nvda are optimized for rapid speech.  Some time ago, a friend of mine and I developed one just for that purpose.  We called it bullet.  We did not share it with anyone because it served our purposes and we did not have the time to beta test it on a wide variety of systems or with languages other than english.

 

We live in the U.S. We use the english (America) voice.  We tested it with all of the english voices.

 

If you are an english speaker and you can stand the english (america) voice, we would be willing to share our Bullet epeak variant with you.

 

I usually run at 40% with rate boost checked.  I set pitch to 40 and inflection to 75.

 

I can’t give you an estimate on words per minute but it is several hundred.

 

I have tested it up to 70% with rate boost checked and it doesn’t seem to distort.

 

I listen with headphones or a set of wired speakers with a separate woofer but the variant sounds good on my laptop speakers as well.

 

Let me know if you would like to try it out.

 

George

 

From: Sociohack AC

Sent: Tuesday, July 17, 2018 2:24 AM

Subject: Re: [nvda] Tips for speed reading /listening with screen readers

 

You get used to it. Words don't run into each other, you can recognize them as you do at slower speeds. Default speeds are too slow and inefficient for navigating around. You can easily go beyond default speeds, how high, that is a question. And, perhaps depends from person to person.
By flutter I mean, a kind of shiver in the voice or the voice getting more robotic than usual. In E-speak it occurs at 100% and at alll boost mode rates. It is still comprehensible, just annoying.
--
Regards,
Sociohack

Re: Tips for speed reading /listening with screen readers

Ervin, Glenn
 

We write it the way we say it.

If someone asks us the date we would say:

June 11

Or November 6.

Glenn

 

From: nvda@nvda.groups.io <nvda@nvda.groups.io> On Behalf Of Steve Nutt
Sent: Tuesday, July 17, 2018 10:34 AM
To: nvda@nvda.groups.io
Subject: Re: [nvda] Tips for speed reading /listening with screen readers

 

Actually, there are other problems with American English which makes me stick with Eloquence for British English.  Americans don’t know how to write dates properly, they are the only country that do it upside down.  So 6/11 is Sixth November in English and 11 June in American.

 

I actually quite like the British version of Eloquence and I am British.

 

There is no difference in intelligibility either.

 

All the best

 

Steve

 

From: nvda@nvda.groups.io <nvda@nvda.groups.io> On Behalf Of Gene
Sent: 17 July 2018 13:44
To: nvda@nvda.groups.io
Subject: Re: [nvda] Tips for speed reading /listening with screen readers

 

I also suspect that in the case of Eloquence, high speed intelligibility is better, perhaps much better, when using American English instead of Brittish English.  I enjoy listening to real Brittish English.  But the Brittish English accent in Eloquence was obviously created by Americans who have no idea how to properly reproduce a Brittish accent.  It's revolting.  Being Americans, they properly reproduced the American accent. 

 

The American accent is probably clearer at fast speeds no matter what accent a person is used to because the Brittish accent degrades the speech itself, it doesn't just change the accent.

 

Gene

----- Original Message -----

From: Sociohack AC

Sent: Tuesday, July 17, 2018 1:24 AM

Subject: Re: [nvda] Tips for speed reading /listening with screen readers

 

I'm so used to E-speak that Eloquence is incoherent to me. But, the flutter you get in E-speak in the boost mode is hindering me to achieve faster speeds. So, I'm forcing myself to get used to Eloquence.

And yes, the so called natural sounding synthesizers are not as good at high speeds. But then, SAPI5 and one core voices aren't much behind if we compare them to these so called natural voices at moderate voices. I tried a demo of Acapella, didn't like it much. It's high quality voices do sound good, and more human like, but you can't use them at high speeds. The speech becomes incoherent. There is a clatter in the background.  When it comes to high speed functionality, there is nothing better than E[speak and Eloquence.
--
Regards,
Sociohack
Re: NVDA vs. Narrator: NVDA still wins.

John Isige
 

Honestly, I'm not knocking them for trying. I don't see myself switching
from NVDA even if Narrator gets to be a viable screen reader. But we
should want it to be and probably get familiar with it, because it's
part of the OS. I'm sure there could be situations where not even NVDA
as a portable copy can be run, so in that case, you'd have a screen
reader that can't be unintentionally blocked by IMO valid security
policies. The more choices we can have, the better.


I should have said, since somebody suggested scan mode, I did get
Narrator to read Firefox technically, in fact what I wanted to test was
jumping by headings in a webpage with 'h' in scan mode. NVDA does that
just fine, e.g. searching on Youtube and hitting 'h' to move to each new
video. Narrator pauses for like a second and says "no next heading".
I'll try it in Edge and Chrome probably, but it also doesn't seem to
read mail in Thunderbird. I guess I could try setting up Microsoft's
mail thingy since all of my accounts are IMAP, but then I feel like I
shouldn't have to put in a ton of work to switch apps, even temporarily,
just to learn another screen reader. If it doesn't work with what I'm
using, that means it doesn't work, as far as I'm concerned.


It was interesting to see the contrast, I can fire up NVDA and it just
works with my stuff, and honestly there's not a whole lot I run into
that it can't handle. I think Microsoft has done a lot with Narrator,
but IMO it's still got a long way to go before it's any sort of viable
alternative. I hope it gets there, and honestly I feel like they're
really trying to work on that, at least so far. It's certainly come a
long way from XP where it didn't really do all that much. But even now,
what a contrast between it and something like NVDA. I will say though,
I'm glad it's adopting keystrokes. The more of a standard we have, the
better. The less I have to go "oh wait, this is Windows and screen
reader X so I have to do B instead of A" means the easier it is for me
to get more stuff done, e.g. if every screen reader using a keyboard has
'h' for moving through headings, then I already know how to navigate
websites. So to my mind, that's actually what we want to see.

On 7/17/2018 6:46, Brian's Mail list account via Groups.Io wrote:
Yews I noticed this hype from Microsoft. they have started visiting
visually impaired charities etc, extolling the virtues of out of the
box accessibility in demos, but you are right, they are always using
Microsoft software. I guess it makes sense for them to get narrator
fast and working with in house stuff before tackling third party
things, and of course one of their latest things is similar  key
sequences to other  screenreaders we all know.
Ahem.
Brian

bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal E-mail to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
----- Original Message ----- From: "John Isige" <gwynn@tds.net>
To: <nvda@nvda.groups.io>
Sent: Tuesday, July 17, 2018 2:21 AM
Subject: [nvda] NVDA vs. Narrator: NVDA still wins.


So I listened to a very nice podcast about one of the new Narrator
features on the Insiders build, the start guide. I fired up Narrator,
just to see what it was like. Opened Firefox, and Narrator pretty
much wouldn't read or navigate a site, Youtube as it happens. Popped
NVDA back on, and it worked just fine.


Not to turn this into a Narrator thread, but does Narrator only work,
or mostly work, with Microsoft apps? I periodically test Narrator
because it's there so why not? Plus Windows is integrating it more,
e.g. you can supposedly install 10 from scratch using Narrator,
Narrator in safe mode, and so on. I figure that makes it worth
getting familiar with if for no other reason than I'll know how to
use it when it's the thing I'll need to use. But it's gonna be hard
to become familiar with it if it doesn't work with chunks of the
stuff I use. Is there a Narrator list or something, like this list
basically? I tried looking around for something like "using Microsoft
narrator with Firefox", but I just get stuff about using Narrator
generally or occasional reviews of using older versions of Narrator.


I'm thankful Microsoft is making accessibility a bigger priority and
giving us things like installation and safe mode, well I've never
really tried either but you know, I'm glad they're there if something
comes up. But for me, Narrator's nowhere near NVDA. I know some
people have implied that it is, and I'm not trying to make this a
fight or anything. So I'm sort of trying to find out, you know does
all of this great functionality come from sticking solely to
Microsoft's stuff? Because I don't get the hype. I can't imagine
using it as my primary screen reader for any length of time. People
say it's getting closer to things like NVDA. But for me, I just don't
see it. Am I missing something? Because right now I think my dream
scenario would be NVDA from power on, or as close as it could be
managed.
Re: Skype and messages or events.

Sarah k Alawami
 

There is a list I run where we are discussing skype a lot, in fact it’s called skype english. Go to skypeenglish.tffppodcast.com for more info and to subscribe.

Take care

On Jul 16, 2018, at 6:06 PM, John Isige <gwynn@...> wrote:

Oh cool! Thanks Joseph. I'll still have to run skype to check periodically, but at least it's fairly easy once it's open.


On 7/16/2018 19:48, Joseph Lee wrote:
Hi,
For the UWP version, pressing Alt+1 to move to conversations will present missed calls and messages in reverse chronological order.
Cheers,
Joseph

-----Original Message-----
From: nvda@nvda.groups.io <nvda@nvda.groups.io> On Behalf Of John Isige
Sent: Monday, July 16, 2018 5:47 PM
To: nvda@nvda.groups.io
Subject: [nvda] Skype and messages or events.

So question about the new Skype, and I think this should apply for 8 or the version in Windows 10. One of the things I really like about Skype classic is having conversations in separate windows. This isn't so much because it's easier to switch conversations, though it is, but rather because of something like this.


Suppose I'm at my machine, and then I say "hey self, I know what we're gonna do, we're gonna go practice mandolin for an hour and clarinet for another hour. Then we'll get lunch"! So I do that, and come back to my machine two and a half hours later. How in the world do I know if somebody tried to call me, or if I got a new chat message? With Skype classic, no problem, I alt-tab through windows and see "oh hey, there's a Skype window open". I'm pretty sure the longest I can tell notifications to stay open is five minutes, and even then, I can't always alt-tab to them, even if they're open. I also don't think they hang around in the action center for that long, do they?


There's probably an easy way to handle this and I just don't know it yet, because I have Skype for 10 on the laptop and I don't do much with Skype there, honestly I haven't done tons with Skype lately for some reason. But I'm sure that will change. Is there a fairly easy to find out if and when new things have happened?
Re: NVDA vs. Narrator: NVDA still wins.

Sarah k Alawami
 

Narrator and chrome actually work a lot better than nva. In fact I do use that more often. It depends on what I need it for. But yeah streamlabs and obs work better with narrator as I can actually read the labels the devs don’t have for nvda. It’s odd.

On Jul 17, 2018, at 4:47 AM, Brian's Mail list account via Groups.Io <bglists@...> wrote:

I'm using water fox inwindows 7, its so much faster toopen pages than Firefox.
Brian

bglists@...
Sent via blueyonder.
Please address personal E-mail to:-
briang1@..., putting 'Brian Gaff'
in the display name field.
----- Original Message ----- From: "Steve Nutt" <steve@...>
To: <nvda@nvda.groups.io>
Sent: Tuesday, July 17, 2018 8:48 AM
Subject: Re: [nvda] NVDA vs. Narrator: NVDA still wins.


Hi,

Try Narrator with Chrome.  I find firefox is officially awful now.

All the best

Steve

-----Original Message-----
From: nvda@nvda.groups.io <nvda@nvda.groups.io> On Behalf Of John Isige
Sent: 17 July 2018 02:22
To: nvda@nvda.groups.io
Subject: [nvda] NVDA vs. Narrator: NVDA still wins.

So I listened to a very nice podcast about one of the new Narrator features on the Insiders build, the start guide. I fired up Narrator, just to see what it was like. Opened Firefox, and Narrator pretty much wouldn't read or navigate a site, Youtube as it happens. Popped NVDA back on, and it worked just fine.


Not to turn this into a Narrator thread, but does Narrator only work, or mostly work, with Microsoft apps? I periodically test Narrator because it's there so why not? Plus Windows is integrating it more, e.g. you can supposedly install 10 from scratch using Narrator, Narrator in safe mode, and so on. I figure that makes it worth getting familiar with if for no other reason than I'll know how to use it when it's the thing I'll need to use. But it's gonna be hard to become familiar with it if it doesn't work with chunks of the stuff I use. Is there a Narrator list or something, like this list basically? I tried looking around for something like "using Microsoft narrator with Firefox", but I just get stuff about using Narrator generally or occasional reviews of using older versions of Narrator.


I'm thankful Microsoft is making accessibility a bigger priority and giving us things like installation and safe mode, well I've never really tried either but you know, I'm glad they're there if something comes up.
But for me, Narrator's nowhere near NVDA. I know some people have implied that it is, and I'm not trying to make this a fight or anything.
So I'm sort of trying to find out, you know does all of this great functionality come from sticking solely to Microsoft's stuff? Because I don't get the hype. I can't imagine using it as my primary screen reader for any length of time. People say it's getting closer to things like NVDA. But for me, I just don't see it. Am I missing something? Because right now I think my dream scenario would be NVDA from power on, or as close as it could be managed.
Re: Beginning NVDA Developer exam: a prototype version

Joseph Lee
 

Hi,

The best place would be the development guide, located at:

https://www.nvaccess.org/files/nvda/documentation/developerGuide.html

Cheers,

Joseph

 

From: nvda@nvda.groups.io <nvda@nvda.groups.io> On Behalf Of Howard
Sent: Tuesday, July 17, 2018 5:55 AM
To: nvda@nvda.groups.io
Subject: Re: [nvda] Beginning NVDA Developer exam: a prototype version

 

Very interesting collection of questions, Joseph! Thank you for putting this list together.

 

For me, it did a good job of pointing out what I do NOT know about N V D A. I am a programmer by profession, but I have not tackled N V D A development. For somebody like me, where is the best place to start to learn how N V D A is structured? Is there a developer's tutorial?

 

On Mon, Jul 16, 2018 at 2:48 PM, Joseph Lee <joseph.lee22590@...> wrote:

Hi all,

 

A few weeks ago on a subgroup of NVDA users list, I announced the availability of an exam that tests people’s readiness to become beginning NVDA developer. A few days after that, I announced that this exam will be made available to the public, and here it is. I recommend test driving this exam (answering all questions) and let me know if you have feedback on this.

 

A prototype certification exam for beginning NVDA developer:

 

Instructions: the questions below will let you meet different facets of NVDA development (both add-ons and Core, although emphasis will be on Core for now). For those seeking to move onto NVDA development should answer ALL questions. Total points will be 100, with passing score being 70 or higher. There will be up to 10 extra credit points, and they will count when calculating scores (if you choose to do them). Answers should be submitted to me Joseph Lee (me) via email no later than July 31, 2018 at 5 PM pacific (8 PM Eastern, 00:00 UTC on August 1st). For some questions, have the NVDA source code in front of you.

 

Additionally, for questions 3 and 4, you can select one of them to be a time-limited question where you have an hour to answer it (be sure to have a stopwatch handy). The timer should begin the moment specified by the question(s). In the answer sheet (email reply), you need to provide how long it took you to answer this question, and without that, five points will be deducted from your score.

 

A special note for folks who took the NVDA Development course: you don't have to take the exam again, and you'll notice slight increase in difficulty with this version.

 

Question 1: a new user of NVDA wants to know the answer to the following questions (write your answer in complete sentences, and answer it as though you’re talking to this user, not me; points: 20, five points per question):

* How does a screen reader work?

* Can you tell me how I can move to a control that can’t be navigated with a keyboard?

* If I press NVDA+Tab, what information will I hear?

* If I found an issue with NVDA, where do I go to report them and what info should I include?

* Extra credit (five points): I want to use my favorite settings before I log onto my computer. How can I do that?

 

Question 2: pick four commands from the following list. For each command, give its command, a short description, and a trace of how the command works. Then if you want, explain in your own words how the command really works. Write it as though you’re producing a video tutorial for power users who wants to know more about how a command works (points: 20, five points per command):

* Read title

* Read focused object

* Read current navigator object

* Move to the next object

* Move to the previous object

* Move navigator object to focus

* Move system focus to the navigator object

* Read foreground window

* Developer info for the navigator object

* Move mouse to current navigator object

* Move between table cells

* Save and revert to saved configuration

* Current app module

* Navigator object location

* Opening NVDA settings panels (any panel will do)

* Review cursor (any cursor command will do)

If you choose to answer settings panel or review cursor commands, choose only one command, as they are a collection of commands. If your answer consists entirely of four commands from one or both of these, you’ll get zero points.

 

The next two questions are “time limited” questions and you need to mark one of them as such. For example, you can mark question 3 as “time limited”, which means question 4 will not be subject to time limits. Once chosen, you’ll have one hour to answer the marked question. The purpose of this exercise is to simulate time limited research (question 3) or programming lab (question 4). If you are adventurous about providing code-based solutions, choose question 4 as time limited question. If you feel writing or research is your strong point, choose question 3 as time limited question. After answering the time limited question, be sure to record how long it took you.

 

Question 3: These days, one of the biggest things software development firms (and projects) look for is ability to research things and communicate with others. To get you to experience this, choose two terms from the list below and write an “internals” (explanatory) paragraph (or two) as though you’re giving a talk at a developers’ conference. Your answer must contain definitions, explanations (in your own words), use cases, examples, and a place in NVDA where it is used. Extra points will be given for one or more of the following additions: actual NVDA code that demonstrates the concept, a reference or two outside of NVDA, and an offer of code or concept redesign to make it useful in the future, or a flaw with current implementations (overall points: 20, ten per item).

If you chose to mark this question as “time limited”, you can just write definitions, use cases, explanations, examples, and if time permits, actual NVDA code examples; no need to include references, design flaws and such, but you’re more than welcome to do so within the time limit (hence extra points). Your time begins once you choose the two items you’d like to write about.

 

* NVDA’s overall architecture

* History of NVDA development and the community

* NVDA object

* Event

* Input and output

* Object hierarchy

* Review cursor

* Object properties

* Text infos

* Browse mode internals

* One accessibility API (choose from MSAA, UIA, or Java Access Bridge)

* Overlay classes

* GUI subsystem

* User configuration

* Add-ons

* NVDA release process

 

Scoring for each term:

* Three points: gave definitions.

* Five points: definitions and explanation in your own words.

* Seven points: definition, explanation, use cases and examples.

* Ten points: definitions, your explanation, use cases, examples, NVDA code example or two.

 

Question 4: when it comes to screen reader development, theory isn’t enough; now it is time to showcase your programming skills, and to do that, choose an NVDA issue on GitHub that you think you can program a solution (quickly if you mark this as “time limited”). Be sure to provide a link to the issue, your rationale for choosing it, a short description, and a basic solution design and code. Extra points if you can demonstrate that your code actually achieves something, complete with tests (points: 25).

If you mark this question as “time limited”, then you DO NEED to have at least a basic solution design and code ready within the time limit. If you chose question 3 as “time limited”, then a solution design will be acceptable; time begins the moment you choose an issue, SO CHOOSE IT WISELY!

 

Scoring:

* Five points: identified an issue you wish to work on, complete with links.

* Ten points: identified and described the issue in your own words.

* Fifteen points: identified, described, and wrote a solution design in your own words.

* Twenty points: identified, described, designed, and programmed a solution in your own words (and your own code).

* Twenty-five points: identified, described, designed, programmed, and tested your solution.

 

Extra credit (eligible if you did ALL of the parts above, including coding and testing):

* Three points: your solution was posted as a pull request.

* Five points: your pull request was reviewed by at least one reviewer, including NV Access people and noted third-party contributors (including I).

 

Question 5: being a developer means more than programming a solution. Users matter, and you need to always remember that you are users also. To help you understand this, I’d like to ask you to serve as a resident tech support engineers for users for the next few days: subscribe to the NVDA users list and answer questions on at least three topics until the end of the exam period. Be sure to provide helpful answers when asked, and show respect to users (points: 15, five per topic).

 

Scoring:

* Zero points: disrespectful answers, or answers that showed no effort from your part to understand users.

* Two points: beginning steps, such as phrasing issues in your own words.

* Three points: showing efforts at trying to reproduce the issue, or if not reproducible, telling users (with respect) that it cannot be done.

* Five points: professional answers regardless of if the solution was found.

 

The following questions are all extra credit and won’t count toward your final score. As they are community activities, your participation is highly encouraged.

 

Question 6 (extra credit, five points): review at least one NVDA add-on marked for review (for details, see NVDA Add-ons mailing list archive).

 

Question 7 (extra credit, five points): test-drive NVDA alpha or beta snapshots throughout the exam period and report bugs if any. Extra points if you can offer solutions.

 

Question 8 (extra credit, five points): find a developer outside of NVDA community that you think can contribute to NVDA in the long run.

 

Question 9 (extra credit, ten points): offer constructive comments on at least two GitHub issues. An extra five points for reviewing a pull request.

 

As noted above, anyone who gets a 70 or higher will be recognized as beginning NVDA developer.

 

P.S. You may have noticed that programming didn’t take the center stage at this exam. The reason: coding alone won’t define you and your talent as a developer. Through this exam I want you all to experience what it is like to be an actual screen reader developer: introducing new users to NVDA, coming up with helpful overviews and tutorials, networking with others and showcasing the project, proposing and programming needed solutions under possible constraints, and providing ongoing technical support for a product.

 

Good luck with the exam.

Cheers,

Joseph



 

--

 

Howard Lee Harkness
918-416-1420
Re: Jeff's addon repository

Sarah k Alawami
 

Your argument has  a flaw. Microsoft will never buy code factory’s stuff. And  for yoru car argument, you would have to see if ther are any patents to do what you want to do, if you wanted to do it that is. And in some countries it is not legal to even do that. One guy got puled over for having a turbo in his ruck he had installed. I think he got fined a big amount of monies for said action.

But as for microsoft or nvda getting eloquence and opensourcing it, that will hapen when hades freezes over.

On Jul 17, 2018, at 4:50 AM, Josh Kennedy <joshknnd1982@...> wrote:

If I had the money, I would personally buy the rites to eloquence and then I would open source it. And then this whole legal argument about it could finally stop. Using it is sort of a gray area though. Because well once again, lets say I purchase sapi5 eloquence. And then I also buy jaws which also has eloquence. So legally I can use sapi5 eloquence, and eloquence as long as it is the one tied to jaws and jaws is running. But if I want to use another free eloquence addon made by a 3rd party which may offer features neither jaws eloquence nor sapi5 eloquence offers, I cannot. Because some legal document says so? How am I hurting the developer? I paid for the jaws license, I paid for the eloquence sapi5 license. They got money from me. Why should they really care if I choose to use the illegal addon especially since I legally purchased not one but multiple sapi5 copies and a copy of jaws which includes eloquence? I could even go out and purchase the official codeFactory addon. But if it turns out to not be as good of quality as the illegal addon, I would use the illegal one. It's the same with a vehicle. Heck we have more rites when it comes to buying cars and trucks than we do with software. If I could see, and if i bought a car or truck to drive around and got my driver's license. Lets say well the truck I bought I want to give it a more powerful engine. I could switch out the engine for a different one, convert it to an electric vehicle if i wish, modify the engine to give it more power, etcetera. Remove the tires and replace them with caterpillar tracks... And in a way I could open source my ideas and show others how to do all this through youtube videos. The problem is that copyright law itself has to change. It is still stuck back in the 1980s or early 1990s. There is another way to hurt if you really want to, the eloquence developers and that would be just don't buy it and use OneCore voices. But some people due to hearing or other issues need eloquence or eloquence works best for them due to the specific qualities of its speech. Perhaps, to solve this problem once and for all, since eloquence is quite old anyway, the United States and UK and Australian governments should get together, purchase all of the rites to eloquence, and then hand those rites off to microsoft under the condition that microsoft build eloquence in as a free tts as sapi5 and oneCore eloquence right into windows, and also maintain the android version so it keeps working. I suppose I will go into feedback hub and suggest this for ease-of-access features. To solve the eloquence legality issue, the cost of eloquence should be spread out where whenever you buy a new computer, which also gives you the rite to use windows, you also have the rite to use eloquence as free tts built into windows10 or downloadable from microsoft's website for windows7 and above. 

Re: Tips for speed reading /listening with screen readers

George McCoy <slr1bpz@...>
 

Greetings,
 
Unfortunately, none of the espeak variants that come with nvda are optimized for rapid speech.  Some time ago, a friend of mine and I developed one just for that purpose.  We called it bullet.  We did not share it with anyone because it served our purposes and we did not have the time to beta test it on a wide variety of systems or with languages other than english.
 
We live in the U.S. We use the english (America) voice.  We tested it with all of the english voices.
 
If you are an english speaker and you can stand the english (america) voice, we would be willing to share our Bullet epeak variant with you.
 
I usually run at 40% with rate boost checked.  I set pitch to 40 and inflection to 75.
 
I can’t give you an estimate on words per minute but it is several hundred.
 
I have tested it up to 70% with rate boost checked and it doesn’t seem to distort.
 
I listen with headphones or a set of wired speakers with a separate woofer but the variant sounds good on my laptop speakers as well.
 
Let me know if you would like to try it out.
 
George
 

From: Sociohack AC
Sent: Tuesday, July 17, 2018 2:24 AM
To: nvda@nvda.groups.io
Subject: Re: [nvda] Tips for speed reading /listening with screen readers
 
You get used to it. Words don't run into each other, you can recognize them as you do at slower speeds. Default speeds are too slow and inefficient for navigating around. You can easily go beyond default speeds, how high, that is a question. And, perhaps depends from person to person.
By flutter I mean, a kind of shiver in the voice or the voice getting more robotic than usual. In E-speak it occurs at 100% and at alll boost mode rates. It is still comprehensible, just annoying.
--
Regards,
Sociohack

Re: possible eloquence solution

JM Casey <crystallogic@...>
 

Good post. I agree with pretty much everything you've said.

Still holding out on purchasing Eloquence for NVDA for some reason. I recently switched back to Eloquence in my JAWS instalation, from the more "natural" voices, and my computer's performance has improved exponentially as a result. I'd almost forgotten how responsive and snappy this synth is. Unfortunately I'm still stuck with Core1 for NVDA for now. The eSpeak comment made me laugh because it seems so true. Not sure I'm one of those who can really get used to that. I was brought up with the Echo synthesizer for Apple in the 80s, which, unless you had it in "flat" mode, always sounded like it was singing, and eSpeak feels like a throwback to those days almost.

-----Original Message-----
From: nvda@nvda.groups.io <nvda@nvda.groups.io> On Behalf Of Damien Garwood
Sent: July 17, 2018 10:17 AM
To: nvda@nvda.groups.io
Subject: Re: [nvda] possible eloquence solution

Hi Josh,
Eloquence? Open-sourced? Owned by Microsoft? You're joking aren't ya?
Seriously though. The sad and sorry fact is, Eloquence, like Keynote, is a very old synthesiser indeed. With that in mind, due to its age, and the possibility that they probably designed it for NASA spacecrafts (Well, the price tag would certainly suggest it), companies whose prices and DRM policies are stuck in the 1980's are highly unlikely to jump to the 21st century and say, "Well, nothing's happening, let's just open source it". They'd be more likely to say, "Over my dead body". All that hard work for nothing, eh?
As for Microsoft owning it? That's a double whammy.
Firstly, unfortunate though the game of life can sometimes be, we have no choice but to face facts. The majority of TTS voices are now leaning towards a more so-called "natural" feel. This includes Microsoft, who have attempted to make all their voices sound natural right from the
SAPI4 days. Sapi5 and OneCore are no exception. Personally, I think it makes it sound choppy, and lag like trash. So then, the only other option will be ESpeak which, while portable, responsive and open-source, as far as usability goes, sounds to me like a trip back to hardware synths of the 1960's.
Secondly, since Microsoft do a good job of butchering everything that they so much as walk near...No. The idea of Microsoft owning something as quality as Eloquence makes me feel sick. Several potential pages of Microsoft-directed rant suppressed, with great difficulty.
Put it this way. If I were any good at self-learning new systems, I would've switched to Windows 10 long ago. As it is, I'm still on 7. And there was me thinking that was too big a change when I was forced to upgrade from XP!
Cheers,
Damien.

On 17/07/2018 01:10 PM, Josh Kennedy wrote:
It seems like unless this problem is solved it will keep coming up. So
I just went into feedback hub and I recommend all of you do the same.
I wrote the following feedback to Microsoft.

Please purchase all rites to CodeFactory Nuance Eloquence TTS. Offer
it as a free downloadable sapi5 and OneCore voice addons under ease of
access in windows10. And on the Microsoft website as sapi5 for
windows7 users. Also please maintain the android version so it keeps
working and perhaps lower the price to $2 or $3. Nuance CodeFactory
licenses are too restrictive. Adding Eloquence to microsoft’s voice
portfolio would benefit those with hearing impairments. So if
Microsoft owned it, whenever you buy a copy of windows or a new pc,
you also pay for the rite to use eloquence on any pc you buy.and if
you don’t want it, you just do not go into ease of access and download
it. Nuance copyright is stuck back in the 80s and early 90s and has to
change. If not, this issue will probably keep cropping up.

Sent from Mail <https://go.microsoft.com/fwlink/?LinkId=550986>; for
Windows 10
Re: Tips for speed reading /listening with screen readers

Rob Hudson
 

Steve Nutt <steve@comproom.co.uk> wrote:
Sorry, you're wrong. There's about 18 variants of male and female voices
for US Google TTS.

Ok, great. I will admit the only place I've heard them is on a phone, running Android lollipop. That one only seemed to have one voice. It sounded like a voice called Pico.
Re: Tips for speed reading /listening with screen readers

Steve Nutt
 

Sorry, you're wrong. There's about 18 variants of male and female voices
for US Google TTS.

All the best

Steve

-----Original Message-----
From: nvda@nvda.groups.io <nvda@nvda.groups.io> On Behalf Of Rob Hudson
Sent: 17 July 2018 15:36
To: nvda@nvda.groups.io
Subject: Re: [nvda] Tips for speed reading /listening with screen readers

Gene <gsasner@ripco.com> wrote:
That's not surprising. I just looked it up and it is a cloud-based
system. It's interesting and I'm going to play with the demo. But in
fairness to Eloquence, it's like comparing a calculator to a supercomputer.

Also, I think there is only one voice for US English in Google TTS. And it
is a female voice. With all deference to the lovely females around the
world, I prefer to listen to a male voice. Purely for hearing loss issues.
Re: Tips for speed reading /listening with screen readers

Steve Nutt
 

Actually it does, but you just need to listen a bit harder.  Comma pauses are shorter than full stop pauses, so I have no problem with this.

 

What it doesn’t do is question mark inflection, which really annoys me.

 

All the best

 

Steve

 

From: nvda@nvda.groups.io <nvda@nvda.groups.io> On Behalf Of Devin Prater
Sent: 17 July 2018 14:34
To: nvda@nvda.groups.io
Subject: Re: [nvda] Tips for speed reading /listening with screen readers

 

The one problem with Google TTS, especially in Android and ChromeOS, is that it inflects commas and periods the same, just ing differently. Eloquence handles commas in all kinds of situations, even, you know, like in clauses, like you know, where it inflects perfectly in that case. Apple voices do a good job of this as well. Yes, though, Google's pronouniation, even of uncommon words, like Blazblue, pronounced Blaze Blue, a Japanese video game franchise. But that doesn't change the fact that Google's inflection doesn't take into account different punctuation. 

--

Re: Tips for speed reading /listening with screen readers

Steve Nutt
 

ESpeak is just horrible, period.

 

All the best


Steve

 

From: nvda@nvda.groups.io <nvda@nvda.groups.io> On Behalf Of Gene
Sent: 17 July 2018 13:46
To: nvda@nvda.groups.io
Subject: Re: [nvda] Tips for speed reading /listening with screen readers

 

Whatever you call it, E-Speak introduces considerable distortion when the speech is speeded up and the distortion starts at not that fast a speed up.  It gets worse as you increase speed. 

 

Gene

----- Original Message -----

Sent: Tuesday, July 17, 2018 1:26 AM

Subject: Re: [nvda] Tips for speed reading /listening with screen readers

 

I don't try to listen at faster speeds than the default. How the hell can you notice a flutter when the speech is so fast the words run together?

 

On 17/07/2018 4:24 PM, Sociohack AC wrote:

I'm so used to E-speak that Eloquence is incoherent to me. But, the flutter you get in E-speak in the boost mode is hindering me to achieve faster speeds. So, I'm forcing myself to get used to Eloquence.

And yes, the so called natural sounding synthesizers are not as good at high speeds. But then, SAPI5 and one core voices aren't much behind if we compare them to these so called natural voices at moderate voices. I tried a demo of Acapella, didn't like it much. It's high quality voices do sound good, and more human like, but you can't use them at high speeds. The speech becomes incoherent. There is a clatter in the background.  When it comes to high speed functionality, there is nothing better than E[speak and Eloquence.
--
Regards,
Sociohack

 
Re: Tips for speed reading /listening with screen readers

Steve Nutt
 

Actually, there are other problems with American English which makes me stick with Eloquence for British English.  Americans don’t know how to write dates properly, they are the only country that do it upside down.  So 6/11 is Sixth November in English and 11 June in American.

 

I actually quite like the British version of Eloquence and I am British.

 

There is no difference in intelligibility either.

 

All the best

 

Steve

 

From: nvda@nvda.groups.io <nvda@nvda.groups.io> On Behalf Of Gene
Sent: 17 July 2018 13:44
To: nvda@nvda.groups.io
Subject: Re: [nvda] Tips for speed reading /listening with screen readers

 

I also suspect that in the case of Eloquence, high speed intelligibility is better, perhaps much better, when using American English instead of Brittish English.  I enjoy listening to real Brittish English.  But the Brittish English accent in Eloquence was obviously created by Americans who have no idea how to properly reproduce a Brittish accent.  It's revolting.  Being Americans, they properly reproduced the American accent. 

 

The American accent is probably clearer at fast speeds no matter what accent a person is used to because the Brittish accent degrades the speech itself, it doesn't just change the accent.

 

Gene

----- Original Message -----

Sent: Tuesday, July 17, 2018 1:24 AM

Subject: Re: [nvda] Tips for speed reading /listening with screen readers

 

I'm so used to E-speak that Eloquence is incoherent to me. But, the flutter you get in E-speak in the boost mode is hindering me to achieve faster speeds. So, I'm forcing myself to get used to Eloquence.

And yes, the so called natural sounding synthesizers are not as good at high speeds. But then, SAPI5 and one core voices aren't much behind if we compare them to these so called natural voices at moderate voices. I tried a demo of Acapella, didn't like it much. It's high quality voices do sound good, and more human like, but you can't use them at high speeds. The speech becomes incoherent. There is a clatter in the background.  When it comes to high speed functionality, there is nothing better than E[speak and Eloquence.
--
Regards,
Sociohack
Re: possible eloquence solution

Steve Nutt
 

Why should they do this?  You’re effectively stopping Code Factory from making money.  Just buy Eloquence and be done with it.  Besides, Microsoft wouldn’t maintain it.

 

All the best


Steve

 

From: nvda@nvda.groups.io <nvda@nvda.groups.io> On Behalf Of Josh Kennedy
Sent: 17 July 2018 13:11
To: nvda@groups.io
Subject: [nvda] possible eloquence solution

 

It seems like unless this problem is solved it will keep coming up. So I just went into feedback hub and I recommend all of you do the same. I wrote the following feedback to Microsoft.

Please purchase all rites to CodeFactory Nuance Eloquence TTS. Offer it as a free downloadable sapi5 and OneCore voice addons under ease of access in windows10. And on the Microsoft website as sapi5 for windows7 users. Also please maintain the android version so it keeps working and perhaps lower the price to $2 or $3. Nuance CodeFactory licenses are too restrictive. Adding Eloquence to microsoft’s voice portfolio would benefit those with hearing impairments. So if Microsoft owned it, whenever you buy a copy of windows or a new pc, you also pay for the rite to use eloquence on any pc you buy.and if you don’t want it, you just do not go into ease of access and download it. Nuance copyright is stuck back in the 80s and early 90s and has to change. If not, this issue will probably keep cropping up.

 

 

Sent from Mail for Windows 10

 
Re: Tips for speed reading /listening with screen readers

Devin Prater
 

Google TTS has many voices. They sound like they're based off of the same voice, but they do have male voices. 
--

Re: Microsoft is Killing Skype Classic on September 1 -

Brian Vogel
 

I decided to play around with the Skype Win10 App [version 12.1815.210.0] this morning and, on the whole, it seems (to me - big proviso there) to be accessible.

I can land on all major controls, most have keyboard shortcuts, and I'm getting back and forth content in a single text messaging thread read back by NVDA.

If anyone wants to play around with Skype in the Win10 App version to try to figure things out let me know.  In combination with Quick Assist to let me see what's happening on the other end it's likely that the interface can be figured out, as I've seen much, much worse in terms of access to controls, lack of labels for same, etc.

--

Brian - Windows 10 Home, 64-Bit, Version 1803, Build 17134  

    A little kindness from person to person is better than a vast love for all humankind.

           ~ Richard Dehmel

 

 
NVDA/EDGE/RSS

Pascal Lambert
 

Hi All,
I have witnessed a lot of improvements in Edge and I am using it more. I
can not figure out how to make use of the RSS for frequent info update on
topics of my choosing. I would appreciate any help/suggestions on this.
Many thanks.
Blessings
Pascal
37601 - 37620 of 81039  
More
Home Messages Hashtags Subgroups About Features Pricing Updates Terms Help Twitter