Voices do not install/run unless you format it exactly like old releases

[thennal10]

The files have to be named in the voice-language-name-quality format, as noted in a #2. Furthermore, the current piper voice configs have some extra elements, such as "dataset", "language", "piper_version", etc. If you don't remove these the voice installs just fine but selecting the voice does not actually run it, the synth just keeps using the old voice (I assume the new voice errors out and it defaults back to the previously used one). It isn't that big of an issue right now as you just need to edit it out but additional configs breaking everything seems to be a bad idea as far as maintainability goes.

[GianlucaApollaro]

Hello. I trained my own voice using Piper, but I can't load it in the addon. I can get it installed, but it just doesn't load. I get an error that says the synth can't be loaded. Is there a guide line on the exact steps to perform to train?