Can I read ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) on EtoBox?

ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) by MacLaughlin, Ansel (author);Rumshisky, Anna (author);Khaziev, Rinat (author);Ramakrishna, Anil (author);Merhav, Yuval (author);Gupta, Rahul (author) is a scholarly article available to read on EtoBox.

What is ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) about?

Negative feedback received from users of voice agents can provide valuable training signal to their underlying ML systems. However, such systems tend to have complex inference pipelines consisting of multiple model-based and deterministic components. Therefore, when negative feedback is received, it can be difficult to attribute the system error to a specific sub-component. In this work, we address this challenge by building a system for error attribution and correction. We prototype attributing errors to the ML models used for domain classification (DC) in the NLU component of an assistant's pipeline, using a combination of a model and rule based system. We propose a simple method to add these detected errors directly to offline DC model training, and study our system's effectiveness on a challenging test set of low-frequency utterances. Our experiments on nine domains suggest that augmenting DC training data with our method significantly improves performance on a majority of them.

Author

MacLaughlin, Ansel (author);Rumshisky, Anna (author);Khaziev, Rinat (author);Ramakrishna, Anil (author);Merhav, Yuval (author);Gupta, Rahul (author)

Publisher

IEEE

Published

2023

Language

EN