Interesting nvda add-on from Tony Gebhard, Interactive NVDA screen reader training add-on — guided, hands-on lessons built into NVDA
Submitted for addon store review, you can check out the Github repo here,
github.com/tonygeb23/nvdaCoach…
Thrilled to share something I've been building: NVDA Coach is now available —
a free, interactive training add-on for the NVDA screen reader. 🎉
As an assistive technology instructor, I've worked with a lot of people who are
just starting out with screen readers. The hardest part isn't finding the right
resource — it's the gap between reading about a command and actually using it.
NVDA Coach closes that gap.
Instead of reading a manual, students practice real NVDA commands step by step,
inside NVDA itself, with spoken instructions and instant feedback.
📚 Version 1.0.0 includes 24 lessons across three chapters:
→ Getting Started — the 10 essential commands every beginner needs
→ Reading and Moving Through Text — character, word, line, and selection navigation
→ Browse Mode and Web Navigation — heading, link, form, and landmark navigation
What makes it different:
✅ Everything happens inside NVDA — no PDFs, no videos, no switching context
✅ Live practice forms for hands-on Tab and control navigation
✅ Practice text embedded directly in the lesson window for reading drills
✅ A practice web page auto-opens for browse mode lessons
✅ Completely free. No account. No internet required after install.
✅ Built for beginners, usable independently or with an instructor
If you work with people who are learning NVDA, or if you're learning yourself,
I'd love for you to try it and share it with anyone who might benefit.
This is a beta release with plenty of room for feedback, comments, and suggestions. Please feel free to flood my email inbox. :)
🔗 Learn more: tonygebhard.me/NVDACoach
📧 Questions or feedback: info@tonygebhard.me
#NVDA #ScreenReader #AssistiveTechnology #Accessibility #AT #BlindTech
#DigitalInclusion #AccessibilityTools #A11y #AssistiveTech
Interactive NVDA screen reader training add-on — guided, hands-on lessons built into NVDA - tonygeb23/nvdaCoach-
GitHub
Jay Pellis
in reply to Leowyatt88 • • •Leowyatt88
in reply to Jay Pellis • • •Timothy Wynn
in reply to Leowyatt88 • • •Thanks for putting this together. I would like a description of all the models, but I suppose web searching will do me good for the more popular ones. They are not categorized by type according to the manual, so I am trying, on a lark, one of the models for instrumental extraction, and hen using the CPU Windows version, I get:
Error: Model file Roformer Model: MelBand Roformer Kim | Inst V2 by Unwa not found in supported model files
Is this because I'm using a CPU?
Leowyatt88
in reply to Timothy Wynn • • •Leowyatt88
in reply to Timothy Wynn • • •Timothy Wynn
in reply to Leowyatt88 • • •Spacedog
in reply to Leowyatt88 • • •Gavin Grundlingh
in reply to Leowyatt88 • • •Leowyatt88
in reply to Gavin Grundlingh • • •Gavin Grundlingh
in reply to Leowyatt88 • • •I'm on Windows. I have the "BS-Roformer-SW.ckpt" file in the models folder, and after I select the input file and output folder, I press the "Start Separation" button. Here is the error I get in the log.
Error: HTTPSConnectionPool(host='raw.githubusercontent.com', port=443): Max retries exceeded with url: /TRvlvr/application_data/main/filelists/download_checks.json (Caused by NameResolutionError("HTTPSConnection(host='raw.githubusercontent.com', port=443): Failed to resolve 'raw.githubusercontent.com' ([Errno 11001] getaddrinfo failed)"))
In addition, an error dialog is displayed with no error text. NVDA just says "Error dialog. OK button."
Is this supposed to happen when there is no Internet connection? I am sometimes offline when wanting to do a separation. I'm running the CPU version of the program.
Leowyatt88
in reply to Gavin Grundlingh • • •x0
in reply to Leowyatt88 • • •