[go: up one dir, main page]
More Web Proxy on the site http://driver.im/

Establishing degrees of closeness between audio recordings along different dimensions using large-scale cross-lingual models

Maxime Fily, Guillaume Wisniewski, Severine Guillaume, Gilles Adda, Alexis Michaud


Abstract
In the highly constrained context of low-resource language studies, we explore vector representations of speech from a pretrained model to determine their level of abstraction with regard to the audio signal. We propose a new unsupervised method using ABX tests on audio recordings with carefully curated metadata to shed light on the type of information present in the representations. ABX tests determine whether the representations computed by a multilingual speech model encode a given characteristic. Three experiments are devised: one on room acoustics aspects, one on linguistic genre, and one on phonetic aspects. The results confirm that the representations extracted from recordings with different linguistic/extra-linguistic characteristics differ along the same lines. Embedding more audio signal in one vector better discriminates extra-linguistic characteristics, whereas shorter snippets are better to distinguish segmental information. The method is fully unsupervised, potentially opening new research avenues for comparative work on under-documented languages.
Anthology ID:
2024.findings-eacl.154
Volume:
Findings of the Association for Computational Linguistics: EACL 2024
Month:
March
Year:
2024
Address:
St. Julian’s, Malta
Editors:
Yvette Graham, Matthew Purver
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
2332–2341
Language:
URL:
https://aclanthology.org/2024.findings-eacl.154
DOI:
Bibkey:
Cite (ACL):
Maxime Fily, Guillaume Wisniewski, Severine Guillaume, Gilles Adda, and Alexis Michaud. 2024. Establishing degrees of closeness between audio recordings along different dimensions using large-scale cross-lingual models. In Findings of the Association for Computational Linguistics: EACL 2024, pages 2332–2341, St. Julian’s, Malta. Association for Computational Linguistics.
Cite (Informal):
Establishing degrees of closeness between audio recordings along different dimensions using large-scale cross-lingual models (Fily et al., Findings 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.findings-eacl.154.pdf
Video:
 https://aclanthology.org/2024.findings-eacl.154.mp4