Unsupervised Subtitle Segmentation with Masked Language Models
Autores:
Fecha: 09.07.2023
Abstract
We describe a novel unsupervised approach to subtitle segmentation, based on pretrained masked language models, where line endings and subtitle breaks are predicted according to the likelihood of punctuation to occur at candidate segmentation points. Our approach obtained competitive results in terms of segmentation accuracy across metrics, while also fully preserving the original text and complying with length constraints. Although supervised models trained on in-domain data and with access to source audio information can provide better segmentation accuracy, our approach is highly portable across languages and domains and may constitute a robust off-the-shelf solution for subtitle segmentation.
BIB_text
title = {Unsupervised Subtitle Segmentation with Masked Language Models},
pages = {771-781},
keywds = {
Audio information; Highly-portable; Language model; Length constraints; Segmentation accuracy; Unsupervised approaches
}
abstract = {
We describe a novel unsupervised approach to subtitle segmentation, based on pretrained masked language models, where line endings and subtitle breaks are predicted according to the likelihood of punctuation to occur at candidate segmentation points. Our approach obtained competitive results in terms of segmentation accuracy across metrics, while also fully preserving the original text and complying with length constraints. Although supervised models trained on in-domain data and with access to source audio information can provide better segmentation accuracy, our approach is highly portable across languages and domains and may constitute a robust off-the-shelf solution for subtitle segmentation.
}
isbn = {978-195942971-5},
date = {2023-07-09},
}
