Homophone Disambiguation Reveals Patterns of Context Mixing in Speech Transformers

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Abstract

Transformers have become a key architecture in speech processing, but our understanding of how they build up representations of acoustic and linguistic structure is limited. In this study, we address this gap by investigating how measures of ‘context-mixing’ developed for text models can be adapted and applied to models of spoken language. We identify a linguistic phenomenon that is ideal for such a case study: homophony in French (e.g. livre vs livres), where a speech recognition model has to attend to syntactic cues such as determiners and pronouns in order to disambiguate spoken words with identical pronunciations and transcribe them while respecting grammatical agreement. We perform a series of controlled experiments and probing analyses on Transformer-based speech models. Our findings reveal that representations in encoder-only models effectively incorporate these cues to identify the correct transcription, whereas encoders in encoder-decoder models mainly relegate the task of capturing contextual dependencies to decoder modules.
Original languageEnglish
Title of host publicationProceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
Place of PublicationSingapore
PublisherAssociation for Computational Linguistics
Pages8249–8260
Number of pages12
DOIs
Publication statusPublished - Dec 2023

Fingerprint

Dive into the research topics of 'Homophone Disambiguation Reveals Patterns of Context Mixing in Speech Transformers'. Together they form a unique fingerprint.

Cite this