Silero Models: pre-trained speech-to-text, text-to-speech and text-enhancement models made embarrassingly simple - View it on GitHub