Speech Synthesis - From Models to Signals and Back - Day 6 Afternoon

Explore the comprehensive field of speech synthesis in this 3-hour 20-minute conference lecture delivered by Associate Professor Jindřich Matoušek from the University of West Bohemia at JSALT 2025. Delve into the evolution of speech synthesis technology through a detailed examination of its historical development, current state-of-the-art methodologies, and emerging future directions. Learn from an expert who has authored over 180 scientific publications and served as Area Chair for Speech Synthesis at major international conferences. Discover how speech synthesis systems transform models into audio signals and the reverse engineering processes involved in understanding and improving these systems. Gain insights into cutting-edge research in computer speech processing, including recent advances in voice banking, speech reconstruction technologies, and applications for accessibility such as solutions for patients after total laryngectomy and language barrier elimination for handicapped viewers. Understand the theoretical foundations and practical implementations that drive modern text-to-speech systems, neural speech synthesis approaches, and the technical challenges involved in creating natural-sounding synthetic speech across different languages and applications.