Vocalizer spells wortds in the NVDA menu instead of saying them again.

[mikebayus]

Steps to reproduce:

Download a Vocalizer expressive TTS voice, go to the NVDA menu by pressing NVDA key plus N, press down arrow once and the vocalizer voice will spell the first word.

Actual behavior:

The vocalizer voices will spell the first word and not just say it.

Expected behavior:

NVDA should say "prefferences"

NVDA logs, crash dumps and other attachments:

System configuration

NVDA installed/portable/running from source:

NVDA version:

NVDA v2024.1 beta1

Windows version:

Windows 11 latest build

Name and version of other software in use when reproducing the issue:

NA

Other information about your system:

NA

Other questions

Does the issue still occur after restarting your computer?

Yes

Have you tried any other versions of NVDA? If so, please report their behaviors.

NVDA v2023.3 doesn't do this.

If NVDA add-ons are disabled, is your problem still occurring?

Yes, I am using Vocalizer Expressive for SAPI5 Screen readers.

Does the issue still occur after you run the COM Registration Fixing Tool in NVDA's tools menu?

I don't know, I never tried that.

[CyrilleB79]

Hello @mikebayus

What happens if you uncheck "Use spelling functionality if supported" in NVDA speech settings?

[mikebayus]

Vocalizer expressive voices say instead of spell.

Great workaround.

On Tue, Dec 19, 2023 at 4:19 AM Cyrille Bougot @.***> wrote:

Hello @mikebayus https://github.com/mikebayus

What happens if you uncheck "Use spelling functionality if supported" in NVDA speech settings?

— Reply to this email directly, view it on GitHub https://github.com/nvaccess/nvda/issues/15940#issuecomment-1862395800, or unsubscribe https://github.com/notifications/unsubscribe-auth/AKJ77NXM7IHSO6KYTR4EMADYKFL3FAVCNFSM6AAAAABA2SMTWGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQNRSGM4TKOBQGA . You are receiving this because you were mentioned.Message ID: @.***>

[CyrilleB79]

OK. So that's probably the same issue as the one with Code Factory add-on (see #14900 and this thread about MathCAT).

Their TTS do not seem to support correctly commands to spell characters when they are embedded in a broader speech sequence, either with the add-on or with the SAPI5 interface.

You'd probably report the issue to them, e.g. with a link to this ticket. Unfortunately, it seem that they do not answer very quickly (according to this thread).

[mikebayus]

I believe it's not that they don't answer very quickly, I think they have no idea what we are talking about.

My other workaround is to default my NVDA Screen Reader to one of my CereProc voices.

I have the Piper TTS addon, or rather I had, until this beta version of NVDA broke it.

It's the same with Acapela until that addon gets updated as well.

I really really really really like the Piper TTS addon, and I am going to go check to see if there is an update just as soon as I send this.

Best, Mike

On Tue, Dec 19, 2023 at 7:07 AM Cyrille Bougot @.***> wrote:

OK. So that's probably the same issue as the one with Code Factory add-on (see #14900 https://github.com/nvaccess/nvda/pull/14900 and this thread about MathCAT https://nvda.groups.io/g/nvda/topic/102523473#112998).

Their TTS do not seem to support correctly commands to spell characters when they are embedded in a broader speech sequence, either with the add-on or with the SAPI5 interface.

You'd probably report the issue to them, e.g. with a link to this ticket. Unfortunately, it seem that they do not answer very quickly (according to this thread https://nvda.groups.io/g/nvda/topic/103075298#113719).

— Reply to this email directly, view it on GitHub https://github.com/nvaccess/nvda/issues/15940#issuecomment-1862640419, or unsubscribe https://github.com/notifications/unsubscribe-auth/AKJ77NXYH3YJNRJVHW4FKRTYKF7RPAVCNFSM6AAAAABA2SMTWGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQNRSGY2DANBRHE . You are receiving this because you were mentioned.Message ID: @.***>

[seanbudd]

Closing - this is an issue with the add-on. Please report it to the add-on authors.

[mikebayus]

You might note that I said earlier in this thread that I don't have the Vocalizer for NVDA Add-on, The Vocalizer expressive driver I have is a SAPI5 Synth.

Coad Factory is non supportive and has no idea what I am talking about, and has no desire to help.

This issue of which I speak was fixed in 2023.2 and is broken again in 2023.1 Beta 2.

Actually, it is 2024.1 Beta 1 but I have since updated.

So if you devs want to close this issue all well and good.

I'll just either default to another synth that works, or unchecked the use spelling functionallity checkbox as a work around.

Happy new year.

Cheers.

On Mon, Jan 1, 2024 at 6:03 PM Sean Budd @.***> wrote:

Closed #15940 https://github.com/nvaccess/nvda/issues/15940 as not planned.

— Reply to this email directly, view it on GitHub https://github.com/nvaccess/nvda/issues/15940#event-11366342288, or unsubscribe https://github.com/notifications/unsubscribe-auth/AKJ77NQJSURRCSLD2VN6VSDYMM6D3AVCNFSM6AAAAABA2SMTWGVHI2DSMVQWIX3LMV45UABCJFZXG5LFIV3GK3TUJZXXI2LGNFRWC5DJN5XDWMJRGM3DMMZUGIZDQOA . You are receiving this because you were mentioned.Message ID: @.***>