Pitch Accent Detection improves Pretrained Automatic Speech Recognition

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sasu, David, Schluter, Natalie
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909726624710656
author Sasu, David
Schluter, Natalie
author_facet Sasu, David
Schluter, Natalie
contents We show the performance of Automatic Speech Recognition (ASR) systems that use semi-supervised speech representations can be boosted by a complimentary pitch accent detection module, by introducing a joint ASR and pitch accent detection model. The pitch accent detection component of our model achieves a significant improvement on the state-of-the-art for the task, closing the gap in F1-score by 41%. Additionally, the ASR performance in joint training decreases WER by 28.3% on LibriSpeech, under limited resource fine-tuning. With these results, we show the importance of extending pretrained speech models to retain or re-learn important prosodic cues such as pitch accent.
format Preprint
id arxiv_https___arxiv_org_abs_2508_04814
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Pitch Accent Detection improves Pretrained Automatic Speech Recognition
Sasu, David
Schluter, Natalie
Computation and Language
Sound
Audio and Speech Processing
We show the performance of Automatic Speech Recognition (ASR) systems that use semi-supervised speech representations can be boosted by a complimentary pitch accent detection module, by introducing a joint ASR and pitch accent detection model. The pitch accent detection component of our model achieves a significant improvement on the state-of-the-art for the task, closing the gap in F1-score by 41%. Additionally, the ASR performance in joint training decreases WER by 28.3% on LibriSpeech, under limited resource fine-tuning. With these results, we show the importance of extending pretrained speech models to retain or re-learn important prosodic cues such as pitch accent.
title Pitch Accent Detection improves Pretrained Automatic Speech Recognition
topic Computation and Language
Sound
Audio and Speech Processing
url https://arxiv.org/abs/2508.04814