Correcting speech pattern


Cearbhall O'Meadhra
 

Hi,

 

How can I stop NVDA from saying names of American states instead of letter combinations such as “Florid” for fl10, or “Nevada” for “NV-Access”?

 

All the best,

 

Cearbhall

 

m +353 (0)833323487 Ph: _353 (0)1-2864623 e: cearbhall.omeadhra@...

 

 




Avast logo

This email has been checked for viruses by Avast antivirus software.
www.avast.com



Patrick Le Baudour
 

Hi,

Unfortunately, it is not a feature of nvda but one specific to the engine/voice, thus there is no simple way to avoid that kind of things altogether.
It should be possible to use the dictionary to replace them one by one, maybe even to find one regexp to fit them all, but to be honest that seems a lot of work for a limited improvement.

-- Patrick

Le 04/06/2021 à 14:49, Cearbhall O'Meadhra a écrit :
Hi,
How can I stop NVDA from saying names of American states instead of letter combinations such as “Florid” for fl10, or “Nevada” for “NV-Access”?
All the best,
Cearbhall
m +353 (0)833323487 Ph: _353 (0)1-2864623 e: cearbhall.omeadhra@blbc.ie
------------------------------------------------------------------------
Avast logo <https://www.avast.com/antivirus>
This email has been checked for viruses by Avast antivirus software.
www.avast.com <https://www.avast.com/antivirus>
<#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>


 

Patrick is absolutely correct that this is a function of the synthesizer(s) in use rather than NVDA itself.  When faced with an unpronounceable sequence like FL10 (F L 1 0) it's got to "take a guess" and since FL is a state abbreviation it makes at least some sense to try "Florida ten."  In the case of NV-Access or NV Access it makes perfect sense to "guess" that NV would be Nevada.

The NV-Access or NV Access case is easy enough as far as a regex match with substitution.  FL10 would not probably require a regex, but that really depends on exactly what it is you're trying to change and how many variants on the pronunciation you need.
--

Brian - Windows 10, 64-Bit, Version 21H1, Build 19043  

I do not understand why some seek to separate a person from their actions.  The self is composed of an individual’s thoughts, actions, and expression, which are contained in and actuated by the body.  What you do and say is the clearest indicator of who you are.

      ~ Brian Vogel

 


 

On Fri, Jun 4, 2021 at 08:50 AM, Cearbhall O'Meadhra wrote:
How can I stop NVDA from saying names of American states instead of letter combinations
-
What's perversely funny is that the OneCore voice, Microsoft David, pronounces these letter sequences as actual letter sequences, not as US state names.  I actually wasn't expecting that to be the case with an American English synth.
--

Brian - Windows 10, 64-Bit, Version 21H1, Build 19043  

I do not understand why some seek to separate a person from their actions.  The self is composed of an individual’s thoughts, actions, and expression, which are contained in and actuated by the body.  What you do and say is the clearest indicator of who you are.

      ~ Brian Vogel

 


Gene
 

If you type the letters themselves, such as fl, in the speech dictionary as the pattern, and then the same letters again in the pronounced as field, you will accomplish what you want. 
 
Gene

-----Original Message-----
Sent: Friday, June 04, 2021 9:48 AM
Subject: Re: [nvda] Correcting speech pattern
 
Patrick is absolutely correct that this is a function of the synthesizer(s) in use rather than NVDA itself.  When faced with an unpronounceable sequence like FL10 (F L 1 0) it's got to "take a guess" and since FL is a state abbreviation it makes at least some sense to try "Florida ten."  In the case of NV-Access or NV Access it makes perfect sense to "guess" that NV would be Nevada.

The NV-Access or NV Access case is easy enough as far as a regex match with substitution.  FL10 would not probably require a regex, but that really depends on exactly what it is you're trying to change and how many variants on the pronunciation you need.
--

Brian - Windows 10, 64-Bit, Version 21H1, Build 19043  

I do not understand why some seek to separate a person from their actions.  The self is composed of an individual’s thoughts, actions, and expression, which are contained in and actuated by the body.  What you do and say is the clearest indicator of who you are.

      ~ Brian Vogel

 


Luke Davis
 

(Sorry this is a few days late; I wrote it then didn't send for some reason)

You can create a regular expression like this in the voice dictionary, which might help:

Case sensitive: yes
Type: regular expression
Pattern: \b(AL|FL|MS|PA)\b
Replacement: \1

Of course, list all 50 states in the pattern.

That will save you from having to maintain 50 different dictionary entries.

Although, test it first with a minimal case, before trying to enter (or find and paste) all the states, to make sure it actually works with your particular synth.

Luke