Using Balabolka with NVDA to edit text

By the-blind-tech-duck, 22 February, 2026

Forum
Windows

Hi all,

We use Balabolka quite often to convert text to audio with JAWS. When doing the conversion, we often need to manually edit URLs or acronyms so they are pronounced correctly. We play the text, listen, stop, edit if needed, then play it again, and so on.

However, I’ve noticed that editing with NVDA is almost impossible without a Braille display. When moving the cursor in the main document field, NVDA says “blank” or “line feed,” and it does not properly announce characters or the current line of text. I tried navigating with the review cursor instead, but it is even worse. Selecting text is almost impossible.

Has anyone found a way to ensure NVDA behaves correctly when editing text in Balabolka? Copying text back and forth between Balabolka and Notepad is not really an option here, given the complexity of the documents. This situation prevents someone who does not have access to JAWS from completing their work and creates employment barriers.

Any suggestions to fix this issue? If not, are you aware of any other software that can convert text to audio that is compatible with NVDA, includes a pronunciation dictionary, and supports SAPI5 voices?

Thanks,

Options

Comments

By Ambro on Tuesday, February 24, 2026 - 04:25

Hello, I am the author of the program Sonarpad, a modern notepad that allows you to open different types of files, convert text to audio using Microsoft voices, SAPI5 and SAPI4, and many other functions, such as reading news, playing podcasts, and importing articles from Wikipedia. The program is written in Rust, which keeps it fast and lightweight.
The program also has a dictionary, accessible from Tools, so you can make replacements that will be used both in reading (F5) and in creating audiobooks. If you want to try it, the links are these:
Portable https://github.com/Ambro86/Sonarpad/releases/latest/download/sonarpad.exe
Installer https://github.com/Ambro86/Sonarpad/releases/latest/download/sonarpad_x64-setup.exe
Cheers,
Ambro

By TheBlindGuy07 on Tuesday, February 24, 2026 - 17:58

I have never tried the windows version, but the dev is very blind / a11y aware and I have heard great things about sc on windows too. Plus on iphone it's a very, very good app.
PS: dear balabolka...

By the-blind-tech-duck on Wednesday, February 25, 2026 - 04:42

Thank you for providing me potential alternatives to Balabolka, this is much appreciated!

Speech Central – Test Feedback

I tested this software, and unfortunately this is not a viable solution in its current state.

It is not possible to edit within the same interface used for reading. Once modifications are saved, the file cannot be reopened in reading mode unless the program is completely closed and relaunched. That workflow is quite inconvenient.

There is also no keyboard shortcut to switch to edit mode, which would significantly improve usability.

Regarding pronunciation features, I noticed a “Pronunciation” button in the settings, but it does not appear to do anything. There also seems to be no pronunciation dictionary available.

Audio export appears to be limited when using SAPI voices — export only works with Microsoft voices. This may be a limitation of the demo version, but it is unclear. Overall, in its current form, the solution is not functional enough for practical use.

SonarPad Feedback

SonarPad is very good — quite similar to Balabolka — and so far the experience has been very positive. It’s also available in French, which is a nice touch.

Here are some points that could be improved:

  • Move cursor during reading: This feature does not seem to work properly. When pressing F6 or F4, the cursor remains on the line where reading originally started instead of following the reading. - Split reading on new lines: Either the documentation is incomplete, or the feature does not behave as expected. I assumed it would pause reading when encountering an empty line, but that does not seem to happen.
  • Pronunciation dictionary: The current implementation feels somewhat limited. It would be helpful to add options allowing:
    • Case-sensitive vs case-insensitive replacement
    • Replacement of whole words only vs partial-word matches
    Additionally, the pronunciation dictionary should be separated from general settings to make exporting and sharing easier.
  • Record Audio Book: When the interface language is set to French, backslashes are automatically appended at the end of the file name, which causes an export error. Manually removing the “\” fixes the issue.
  • Batch export vs Record Audio Book: Batch export allows WAV output, but Record Audio Book does not. The Record Audio Book feature should also allow selecting WAV format, sample rate, and bitrate. For now, I can easily convert files to WAV once generated.
  • Test microphone option: In the Record Podcast dialog, I noticed that the “Test microphone” feature appears broken. It generates very strange noises instead of normal monitoring feedback. This is not a big deal, I noticed it during my tests.

On a positive note, the accessible terminal option is an interesting and promising approach.

If helpful, I would be happy to open detailed bug reports on GitHub.

Thank you again for the great suggestion, and do not hesitate to contact me if you have any questions!