Saved in:
Bibliographic Details
Main Authors: Bentum, Martijn, Bosch, Louis ten, Lentz, Tomas O.
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2507.04738
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916829923901440
author Bentum, Martijn
Bosch, Louis ten
Lentz, Tomas O.
author_facet Bentum, Martijn
Bosch, Louis ten
Lentz, Tomas O.
contents In this paper we study word stress representations learned by self-supervised speech models (S3M), specifically the Wav2vec 2.0 model. We investigate the S3M representations of word stress for five different languages: Three languages with variable or lexical stress (Dutch, English and German) and two languages with fixed or demarcative stress (Hungarian and Polish). We train diagnostic stress classifiers on S3M embeddings and show that they can distinguish between stressed and unstressed syllables in read-aloud short sentences with high accuracy. We also tested language-specificity effects of S3M word stress. The results indicate that the word stress representations are language-specific, with a greater difference between the set of variable versus the set of fixed stressed languages.
format Preprint
id arxiv_https___arxiv_org_abs_2507_04738
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Word stress in self-supervised speech models: A cross-linguistic comparison
Bentum, Martijn
Bosch, Louis ten
Lentz, Tomas O.
Computation and Language
Artificial Intelligence
In this paper we study word stress representations learned by self-supervised speech models (S3M), specifically the Wav2vec 2.0 model. We investigate the S3M representations of word stress for five different languages: Three languages with variable or lexical stress (Dutch, English and German) and two languages with fixed or demarcative stress (Hungarian and Polish). We train diagnostic stress classifiers on S3M embeddings and show that they can distinguish between stressed and unstressed syllables in read-aloud short sentences with high accuracy. We also tested language-specificity effects of S3M word stress. The results indicate that the word stress representations are language-specific, with a greater difference between the set of variable versus the set of fixed stressed languages.
title Word stress in self-supervised speech models: A cross-linguistic comparison
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2507.04738