Naming Conventions in the NVDA Voice Dialogue


Nermin
 

Hi,


maybe this is something Joseph Lee can shed a light on. Ever since I've had to switch synthesizers in NVDA, and I've been using the software since 2007 or so, I've wondered why "Variant" is variant, and "voice" is voice.


To explain this a bit, the "Voice" combo box will effectively switch the synthesiser language, while "variant" switches the current voice.

So, for consistency, wouldn't it be better to just rename them?

I'd assume that the naming convention came into being when ESpeak gained popularity and Jonathan Duddington was still on the then NVDA list.


Regards,

Nermin


Quentin Christensen
 

I'm not familiar with the original decisions as to the naming, however, the exact options do depend on the synthesizer in use.
eSpeak NG uses both options.  I currently have:
Voice: English (Great Britain)
Variant: Max

OneCore only uses voice, which is currently Microsoft George on mine.  Although the language isn't specified, each language includes usually two voices (Male and female) - so I could switch between male and female voices of the one language, or change languages entirely by altering the "voice" here.

Microsoft SPeech API Version 5.  Works similar to OneCore, although the list does include mention of the language of each voice, eg: Microsoft David - English (United States)

I haven't double checked other synthesizers just now, but my understanding is that some synthesizers will have options to alter one or the other.

Quentin.



On Wed, May 4, 2022 at 12:38 PM Nermin via groups.io <voy44=yahoo.de@groups.io> wrote:
Hi,


maybe this is something Joseph Lee can shed a light on. Ever since I've
had to switch synthesizers in NVDA, and I've been using the software
since 2007 or so, I've wondered why "Variant" is variant, and "voice" is
voice.


To explain this a bit, the "Voice" combo box will effectively switch the
synthesiser language, while "variant" switches the current voice.

So, for consistency, wouldn't it be better to just rename them?

I'd assume that the naming convention came into being when ESpeak gained
popularity and Jonathan Duddington was still on the then NVDA list.


Regards,

Nermin








--
Quentin Christensen
Training and Support Manager


David Goldfield
 

I do agree with Nermin and, while I don’t necessarily consider this to be the most important issue to consider, renaming some of these options might make NVDA’s interface to be a bit more intuitive to newer users. I think that the variant option might be more understandable if it could be called Person or Voice with the Voice list being renamed something like Language? I may do a search for this on the Github page and file it as a feature request if it hasn’t already been done.

 

David Goldfield,

Blindness Assistive Technology Specialist

NVDA Certified Expert

 

Subscribe to the Tech-VI announcement list to receive emails regarding news and events in the blindness assistive technology field.

Email: tech-vi+subscribe@groups.io

www.DavidGoldfield.org

 

 

 

 

From: nvda@nvda.groups.io <nvda@nvda.groups.io> On Behalf Of Quentin Christensen
Sent: Tuesday, May 3, 2022 11:33 PM
To: nvda@nvda.groups.io
Subject: Re: [nvda] Naming Conventions in the NVDA Voice Dialogue

 

I'm not familiar with the original decisions as to the naming, however, the exact options do depend on the synthesizer in use.

eSpeak NG uses both options.  I currently have:

Voice: English (Great Britain)

Variant: Max

 

OneCore only uses voice, which is currently Microsoft George on mine.  Although the language isn't specified, each language includes usually two voices (Male and female) - so I could switch between male and female voices of the one language, or change languages entirely by altering the "voice" here.

 

Microsoft SPeech API Version 5.  Works similar to OneCore, although the list does include mention of the language of each voice, eg: Microsoft David - English (United States)

 

I haven't double checked other synthesizers just now, but my understanding is that some synthesizers will have options to alter one or the other.

 

Quentin.

 

 

 

On Wed, May 4, 2022 at 12:38 PM Nermin via groups.io <voy44=yahoo.de@groups.io> wrote:

Hi,


maybe this is something Joseph Lee can shed a light on. Ever since I've
had to switch synthesizers in NVDA, and I've been using the software
since 2007 or so, I've wondered why "Variant" is variant, and "voice" is
voice.


To explain this a bit, the "Voice" combo box will effectively switch the
synthesiser language, while "variant" switches the current voice.

So, for consistency, wouldn't it be better to just rename them?

I'd assume that the naming convention came into being when ESpeak gained
popularity and Jonathan Duddington was still on the then NVDA list.


Regards,

Nermin






 

--

Quentin Christensen
Training and Support Manager

 


 

Hi all,

Since someone did request something from me...

As folks found out, different speech synthesizers organize their voices and variants differently. This made sense back when eSpeak NG was the primary speech synthesizer; this is no longer the case in Windows 10 and later where Windows OneCore is used as default speech synthesizer. Also, remember that some synthesizers combine voice and variatns under the "voice" list.

While I understand the need to change user interface messages to reflect more diverse synthesizer options, we also need to acknowledge how we ended up with what we have. We also need to keep in mind that comparing speech synthesizers is akin to apples to oranges comparison, more so when we think about voice data, subtle differences in pronunciation rules, internal speech engine and model differences, and how speech output parameters (that's what they are actually called in the development circles) are organized and presented both internally and externally. Because this forum had many discussions surrounding specific synthesizers (including licensing and authorship), suffice to say that changing labels for sake of convenience is something that should not be suggested lightly (I hope to use messages like this as teaching moments so we can "take off the mystery mask" surrounding screen reader internals; yes, I'm a fan of a certain South Korean TV show, but won't elaborate further).

Cheers,

Joseph