Saved in:
Bibliographic Details
Main Authors: Shim, Ryan Soh-Eun, De Cristofaro, Domenico, Hu, Chengzhi Martin, Vietti, Alessandro, Plank, Barbara
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.19606
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918426862157824
author Shim, Ryan Soh-Eun
De Cristofaro, Domenico
Hu, Chengzhi Martin
Vietti, Alessandro
Plank, Barbara
author_facet Shim, Ryan Soh-Eun
De Cristofaro, Domenico
Hu, Chengzhi Martin
Vietti, Alessandro
Plank, Barbara
contents Cross-lingual alignment in pretrained language models enables knowledge transfer across languages. Similar alignment has been reported in Whisper-style speech encoders, based on spoken translation retrieval using representational similarity. However, prior work does not control for phonetic overlap between equivalent utterances, which may artificially support retrieval. We conduct pronunciation-controlled experiments to test whether cross-lingual alignment arises from semantic rather than phonetic similarity. Results show that spoken translation retrieval remains strongly above chance without phonetic cues in the final layers of encoders trained with a speech translation objective, most clearly for models additionally trained on translation. We further test early-exiting the encoder to induce representations we hypothesize to be less tied to language-specific semantics. These experiments indeed reveal performance gains in automatic speech recognition on low-resource languages unseen during training.
format Preprint
id arxiv_https___arxiv_org_abs_2505_19606
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Languages in Whisper-Style Speech Encoders Align Both Phonetically and Semantically
Shim, Ryan Soh-Eun
De Cristofaro, Domenico
Hu, Chengzhi Martin
Vietti, Alessandro
Plank, Barbara
Computation and Language
Cross-lingual alignment in pretrained language models enables knowledge transfer across languages. Similar alignment has been reported in Whisper-style speech encoders, based on spoken translation retrieval using representational similarity. However, prior work does not control for phonetic overlap between equivalent utterances, which may artificially support retrieval. We conduct pronunciation-controlled experiments to test whether cross-lingual alignment arises from semantic rather than phonetic similarity. Results show that spoken translation retrieval remains strongly above chance without phonetic cues in the final layers of encoders trained with a speech translation objective, most clearly for models additionally trained on translation. We further test early-exiting the encoder to induce representations we hypothesize to be less tied to language-specific semantics. These experiments indeed reveal performance gains in automatic speech recognition on low-resource languages unseen during training.
title Languages in Whisper-Style Speech Encoders Align Both Phonetically and Semantically
topic Computation and Language
url https://arxiv.org/abs/2505.19606