🇨🇦Samuel Proulx🇨🇦Today I learned why Sonata #TTS created a framework to run AI voices outside of #NVDA. First, NVDA doesn't come with all of the #Python standard libraries. Second, there's no good way of updating dependencies in a bundled addon. Third, NVDA really, really hates it if you include several hundred dependencies in your addon. Anyway, here's kitten TTS, the other synthesizer I wanted to try with NVDA. Unfortunately, the model doesn't support streaming output, so even though it's actually faster than Supertonic, it feels slower when used with NVDA. Also, it takes several minutes to install, makes NVDA startup 30 seconds slower, and freezes the change synthesizer dialogue for about 45 seconds when you open it. It does miss words less frequently, though, and pronounces text better. The ultimate result of my two-day investigation is that even the highly optimized open-source AI imbedded models are not yet ready for screen reader use. Some tree-shaking could fix some of these issues, but it still won't allow for streaming, so it's not worth it. I'd really like to know what Microsoft and Narrator are doing to get the natural voices so snappy. github.com/fastfinge/kittentts-nvda/ #screenreader
Amir
@amir@dragonscave.space
- 241 Followers
- 143 Following
- 353 Posts
Visually impaired interpreter, language instructor, blogger, podcast host/producer, and assistive technology advocate, also an AccessibleAndroid.com writer. Avocations: mountain-climbing and piano-playing -- currently own a Yamaha CLP-745. Situated in Tehran at the moment.
Okay! If you're running 64-bit #NVDA, and you want to play with a new #tts engine that sort of works most of the time, here's the first release. The engine only supports English. github.com/fastfinge/supertonic-nvda/releases/tag/v1
Did some work on a new high quality #nvda #tts engine for 2026. Unfortunately, I can’t get indexing to work. Even with AI. Help? Someone? It builds with scons. GitHub - fastfinge/supertonic-nvda: supertonic for nvda github.com/fastfinge/supertonic-nvda
Based on the information from the Sight and Sound Technology page about the "JAWS error VCSPEECH.JLS is missing" - link at the end, the suggested solution — if it even works — is to remove all Microsoft Visual C++ runtime components from the system so that the JAWS installer can reinstall the specific versions it needs.
But we have to ask: why wasn’t this issue present in the initial JAWS 2026 release and suddenly appeared in the December build? And if we remove all the C++ components, what will happen to other programs that depend on them?
We should also ask why Sight and Sound Technology has offered a workaround for this issue, yet we see nothing about it on Freedom Scientific’s own website. For a piece of expensive software, the tech support response feels amateurish and almost comically unacceptable.
https://sastltd.zohodesk.eu/portal/en/kb/articles/jaws-error-vcspeech-jls-is-missing
But we have to ask: why wasn’t this issue present in the initial JAWS 2026 release and suddenly appeared in the December build? And if we remove all the C++ components, what will happen to other programs that depend on them?
We should also ask why Sight and Sound Technology has offered a workaround for this issue, yet we see nothing about it on Freedom Scientific’s own website. For a piece of expensive software, the tech support response feels amateurish and almost comically unacceptable.
https://sastltd.zohodesk.eu/portal/en/kb/articles/jaws-error-vcspeech-jls-is-missing
JAWS – Error VCSpeech.jls is missing
Problem During launch JAWS posts an error indicating that vcspeech.jls is missing, no speech is heard JAWS fails to Speak.. Applie to JAWS Resolution After installing JAWS you encounter the vcspeech.jls error. This error occurs when JAWS starts, it ...
When JAWS goes crazy!
Both my work desktop and my laptop have JAWS 2026, with the company covering the licenses. But after updating my laptop to the December build of JAWS 2026 a couple of days ago, I started getting this awful error: “cannot load VCSPEECH.JLS.”
Repairing JAWS or even uninstalling and reinstalling it doesn’t fix anything.
I did some online research, and people who ran into the same problem say the only solution is reinstalling Windows 11. Seriously? That’s absolutely ridiculous and completely unacceptable.
Luckily, NVDA is my main screen reader. I’m honestly fed up with trying to troubleshoot this issue. I’m even considering ditching JAWS entirely at work, despite the fact that it still performs better in some environments and scenarios.
Both my work desktop and my laptop have JAWS 2026, with the company covering the licenses. But after updating my laptop to the December build of JAWS 2026 a couple of days ago, I started getting this awful error: “cannot load VCSPEECH.JLS.”
Repairing JAWS or even uninstalling and reinstalling it doesn’t fix anything.
I did some online research, and people who ran into the same problem say the only solution is reinstalling Windows 11. Seriously? That’s absolutely ridiculous and completely unacceptable.
Luckily, NVDA is my main screen reader. I’m honestly fed up with trying to troubleshoot this issue. I’m even considering ditching JAWS entirely at work, despite the fact that it still performs better in some environments and scenarios.
Accessible AndroidTCL 65V6C Google TV Review https://accessibleandroid.com/tcl-65v6c-google-tv-review/
QuinI went on FSCast to talk about Paperback! "This month, Mitchell Smedley shares his experience hosting his own radio show and serving as president of a college radio station. Quin Gillespie introduces Paperback, a fully accessible ebook and document reader for Windows. Finally, Elizabeth Whitaker highlights upcoming training opportunities from Vispero." https://blog.freedomscientific.com/fscast-265-on-hosting-a-radio-show-reading-documents-and-developing-your-professional-skills/ Thanks @freedomscientific for having me!
FSCast #265. On hosting a radio show, reading documents and developing your professional skills
This month, Mitchell Smedley shares his experience hosting his own radio show and serving as president of a college radio station. Quin Gillespie introduces Paperback, a fully accessible ebook and …
So a new mod of the Natural Voice SAPI Adapter has become available which resolves NVDA 64-bit not being able to expose the true HD quality when Microsoft Ava and Microsoft Andrew offline voices are used. The caveat is increased stuttering even on powerful machines. Grab it directly from here:
https://github.com/user-attachments/files/24298799/NaturalVoiceSAPIAdapter_v0.2.7_x86_x64.Mod.x86_1.46.0.x64_1.47.0.zip
https://github.com/user-attachments/files/24298799/NaturalVoiceSAPIAdapter_v0.2.7_x86_x64.Mod.x86_1.46.0.x64_1.47.0.zip
The Ghost D-Pad and Why “Good Enough” is Failing Android Braille Users https://accessibleandroid.com/the-ghost-d-pad-and-why-good-enough-is-failing-android-braille-users/
Finally figured out why I can’t get the true HD sound out of Microsoft Ava and Andrew via the SAPI Adapter when using NVDA (yes, the modded adapter).
The issue is with the new 64-bit NVDA alphas. Since they no longer support 32-bit, they can’t use the 32-bit part of the SAPI Adapter—the part that actually produces the real HD effect.
That’s why the current official NVDA release (still 32-bit) does deliver true HD with Ava and Andrew when used with the SAPI Adapter.
You can test this yourself using TtsApplication.exe in both the x64 and x86 folders. Only the x86 version produces the true HD effect.
The issue is with the new 64-bit NVDA alphas. Since they no longer support 32-bit, they can’t use the 32-bit part of the SAPI Adapter—the part that actually produces the real HD effect.
That’s why the current official NVDA release (still 32-bit) does deliver true HD with Ava and Andrew when used with the SAPI Adapter.
You can test this yourself using TtsApplication.exe in both the x64 and x86 folders. Only the x86 version produces the true HD effect.