Extracting Speaker and Emotion Information from Self-Supervised Speech Models

Explore the extraction of speaker and emotion information from self-supervised speech models in this 38-minute conference talk by Themos Stafylakis from the Center for Language & Speech Processing (CLSP) at Johns Hopkins University. Delivered as part of JSALT 2023, the 30th edition of the workshop held in Le Mans, France, this presentation delves into cutting-edge techniques for analyzing speech data. Learn about the latest advancements in self-supervised learning applied to speech processing, with a focus on extracting valuable information about speakers and their emotional states. Gain insights into the research conducted at CLSP and its potential applications in various fields of speech technology and natural language processing.

Extracting speaker and emotion information from self-supervised speech models -- Themos Stafylakis