In this tutorial, the first session will introduce the theoretical advantages of large-scale pre-trained foundation models by the universal approximation theory and how to update the large-scale speech and acoustic models effectively using parameter-efficient learning. Next, our second session will introduce how we can do effective cross-modal pre-training of representations across visual, speech, and language modalities, which can be learned without necessarily needing aligned data across modalities and can benefit tasks in individual modalities as well. Finally, our third session will explore different applications on multimedia processing benefited from the pre-training of acoustic and language modelling with benchmark performance.

Amazon organizers: Huck Yang and Shalini Ghosh

Website: https://interspeech2023.org/tutorials/#toggle-id-3

The third edition of the Symposium on Security & Privacy in Speech Communication, focuses on speech and voice through which we express ourselves. As speech communication can be used to command virtual assistants to transport emotion or to identify oneself, the symposium encourages participants to give answers on how we can strengthen security and privacy for speech representation types in user-centric human/machine interaction? The symposium therefore sees that interdisciplinary exchange is in high demand and aims to bring together researchers and practitioners across multiple disciplines – more specifically: signal processing, cryptography, security, human-computer interaction, law, ethics, and anthropology.