Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Serious bugs while using Worldvoice #19

Open
Destranis opened this issue May 22, 2024 · 0 comments
Open

Serious bugs while using Worldvoice #19

Destranis opened this issue May 22, 2024 · 0 comments

Comments

@Destranis
Copy link

Hello.
I'm creating an issue to report some bugs I found in Worldvoice.
Let's start with the facts. I am using a setup with Windows 11, the latest version of NVDA, version: 2024.2 beta 1 ( 2024.2.0.31914).
I am using Worldvoice with RH voice set as the default engine and eSpeak as secondary voice. I'll explain why below.
The first bug I see is that the addon is not able to recognize languages properly. I'd like to use the addon with Hungarian as the default language (since it's my native language) and Russian as the secondary one. All the addon needs to do is to recognize Cyrillic letters and pronounce them as words with RH voice. But it doesn't work. eSpeak as the default voice does its job very well, reads Hungarian like a dream. However, it doesn't recognize when the addon sees cirillic letters, so RH voice doesn't even turn on to read at least something. If I turn off Unicode based language recognition, RH Voice reads everything, even Latin letters with a Russian accent. The only way to get Worldvoice to work as expected is to set RH voice as primary tts and eSpeak as secondary. But this causes even more problems.
Of course, if you set RH voice as the default engine, it will intercept all punctuation, emoticons, and any non-letter symbols. So as I type this text, I hear all the letters in Hungarian, but the spaces, commas, and periods in Russian with RH Voice. This is to be expected, since RH Voice is the main engine from now on. But it doesn't work otherwise.
There's no way to adjust the intonation. I don't know what the default is, but eSpeak's default intonation really bothers me. I used to use it at about 35 and now it sounds like 75-80. Please add a setting for us to be able to change it.
Numbers. As I wrote above, my main engine is RH Voice, Russian, and the secondary is eSpeak, Hungarian. eSpeak doesn't say numbers correctly if there is text before them, or if there is a space between them. A good example of this is 1 524, which should be one thousand fivehundred twenty four, but eSpeak says one fivehundred twenty four, of course in hungarian. Do something with this please.
Punctuation, again. As RH voice being the main engine, even if reading punctuation is set to none, RH voice intercepts all the settings and reads out all, I mean, all the punctuation symbols, in russian. If I add them to unicode exceptions, it becomes just worse, as very long pauses appear while continuous speech.
After changing a few settings, the Worldvoice window freezes. It's impossible to close it and only restarting NVDA works. I can't reproduce it, as I don't understand yet, what causes the problem, but it is one of the settings tab for sure, built in Worldvoice itself. This happened to me while configuring speech, and while choosing locale name and voice, so basically any of the windows inside of the addon's settings can cause this.
Sometimes Worldvoice doesn't read content automatically. There are a few programs, from which I get notifications, which the synthesizer should read automatically (or that's what any other synthesizers do.) Dinamic content reading is turned on, and as I said before, every other synthesizer reads everything well, except Worldvoice.
These are all the bugs I've found so far.
I'm in high hopes I could help out, and with all the feedback, the addon can be developed to be even greater.
This was basically my dream to have such a thing for NVDA, and I'm going to do everything to give you feedback and bug reports, so this thing can be even better!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant