XClose

UCL Psychology and Language Sciences

Home
Menu

Speech Science Forum - Gašper Beguš (UC Berkeley)

03 November 2022, 4:00 pm–5:30 pm

Speech Science Forum Logo

Modeling language from raw speech: A deep learning approach

This event is free.

Event Information

Open to

All

Cost

Free

Organiser

Justin Lo

Abstract:

In this talk, I propose that language can be modeled from raw speech data in a fully unsupervised manner with Generative Adversarial Networks (GANs) and that such modeling has implications both for the understanding of language acquisition and for the understanding of how deep neural networks learn internal representations. I propose an extension of the GAN architecture in which learning of meaningful linguistic units emerges from a requirement that the networks output informative data and which captures both the perception and production principles of human speech. I further propose a technique to identify latent variables in deep convolutional networks that represent linguistically meaningful units in a causal and interpretable way. With this model, we can “wug-test” deep neural networks, analyze how their biases match human learning biases in behavioral experiments, how speech processing in the brain compares to intermediate representations in deep neural networks, how symbolic-like rule-like computation emerges in internal representations, and what GANs’ innovative outputs can teach us about productivity in human language.

About the Speaker

Gašper Beguš

at University of California, Berkeley

More about Gašper Beguš