A multi-voice TTS system trained with an emphasis on quality - View it on GitHub