Speech Vecalign: an Embedding-based Method for Aligning Parallel Speech Documents

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Meng, Chutong, Koehn, Philipp
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909800138276864
author Meng, Chutong
Koehn, Philipp
author_facet Meng, Chutong
Koehn, Philipp
contents We present Speech Vecalign, a parallel speech document alignment method that monotonically aligns speech segment embeddings and does not depend on text transcriptions. Compared to the baseline method Global Mining, a variant of speech mining, Speech Vecalign produces longer speech-to-speech alignments. It also demonstrates greater robustness than Local Mining, another speech mining variant, as it produces less noise. We applied Speech Vecalign to 3,000 hours of unlabeled parallel English-German (En-De) speech documents from VoxPopuli, yielding about 1,000 hours of high-quality alignments. We then trained En-De speech-to-speech translation models on the aligned data. Speech Vecalign improves the En-to-De and De-to-En performance over Global Mining by 0.37 and 0.18 ASR-BLEU, respectively. Moreover, our models match or outperform SpeechMatrix model performance, despite using 8 times fewer raw speech documents.
format Preprint
id arxiv_https___arxiv_org_abs_2509_18360
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Speech Vecalign: an Embedding-based Method for Aligning Parallel Speech Documents
Meng, Chutong
Koehn, Philipp
Computation and Language
We present Speech Vecalign, a parallel speech document alignment method that monotonically aligns speech segment embeddings and does not depend on text transcriptions. Compared to the baseline method Global Mining, a variant of speech mining, Speech Vecalign produces longer speech-to-speech alignments. It also demonstrates greater robustness than Local Mining, another speech mining variant, as it produces less noise. We applied Speech Vecalign to 3,000 hours of unlabeled parallel English-German (En-De) speech documents from VoxPopuli, yielding about 1,000 hours of high-quality alignments. We then trained En-De speech-to-speech translation models on the aligned data. Speech Vecalign improves the En-to-De and De-to-En performance over Global Mining by 0.37 and 0.18 ASR-BLEU, respectively. Moreover, our models match or outperform SpeechMatrix model performance, despite using 8 times fewer raw speech documents.
title Speech Vecalign: an Embedding-based Method for Aligning Parallel Speech Documents
topic Computation and Language
url https://arxiv.org/abs/2509.18360