psibre

Sure, in theory. But what would be the advantage over using the festvox build scripts, etc.?

OK, I see where this is going. In fact we already have internal toolchains that work similarly to the [Gradle-based voicebuilding plugins](/marytts/gradle-marytts-voicebuilding-plugin) but produce HTS-compatible files. Ultimately, the models trained...

Thanks very much for the contribution! I'm not sure what's wrong with the CI build, but there have been a few random failures on buildhive in the past. On the...

Yes, everything under `marytts-lang-cs` would be needed, excluding temp stuff such as `target/` and IDE metadata. If you add it to your fork, I can give more detailed feedback for...

Thank you very much! I've merge your contribution into https://github.com/psibre/marytts/tree/czech for now. I'll test the actual voice asap. Leaving the pull request open for now. Unless something unexpected blocks it,...

I'm afraid the instructions for [publishing a new voice](../wiki/Publishing-a-MARY-TTS-Voice) are woefully outdated and no longer compatible with our development practices five years later... We really need to get the documentation...

Indeed, you're just overwriting the value. You want something like: ```java mary.setAudioEffects("TractScaler(amount:1.5;)+F0Add(f0Add:50.0;)"); ``` Note that not all effects can be combined -- you'll have to play around and see what...

Yeah, that's the kind of crap that Eclipse auto-generated and various contributors couldn't be bothered to implement properly. In fact this specific anti-pattern is exactly what -- even back in...

@proewer Building new languages, and building new *voices* (note the distinction) using audio and text files has been a feature for many years. You do not need to wait for...

The READMEs for the various voice repos should be clear enough. Follow the instructions there and open new issues if you have specific questions or encounter errors. Once you understand...