Gašper Beguš: How Does Artificial Intelligence Learn to Speak? From Syllables to Storytelling …

American Slovenian Education Foundation (ASEF) organizes a new ASEF Speaker Series event. This time dr. Gašper Beguš  will deliver a lecture titled “How does artificial intelligence learn to speak? From syllables to storytelling …”

ABSTRACT

Machine learning and neural networks bring many applications in a variety of fields, from self-driving cars to automatic translation from language to language. We know quite a bit about how exactly learning takes place in neural networks. On the other hand, we understand human speech relatively well. Children begin to learn speech before they are born, showing patterns that are common to all children. What language acquisition can teach us about how artificial intelligence learns and vice versa, what artificial intelligence can teach us about how children learn language will be the main topics of this lecture.

BIO

Gašper Beguš, PhD, is an assistant professor in the Department of Linguistics at the University of California, Berkeley. He completed his doctorate at the prestigious Harvard University, then became an assistant professor and set up a phonology laboratory at the University of Washington. This year, he begins his work as an assistant professor at the University of California, Berkeley, where he has set up a laboratory to deal with speech. In his research, he uses neural networks and behavioral experiments to deepen his understanding of how neural networks and humans learn speech.