Stylespeech: Self-Supervised Style Enhancing with VQ-VAE-Based Pre-Training for Expressive Audiobook Speech Synthesis | IEEE Conference Publication | IEEE Xplore