Predicting Prosodic Prominence from Text with Pre-trained Contextualized Word Representations

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review


In this paper we introduce a new natural language processing dataset and benchmark for predicting prosodic prominence from written text. To our knowledge this will be the largest publicly available dataset with prosodic labels. We describe the dataset construction and the resulting benchmark dataset in detail and train a number of different models ranging from feature-based classifiers to neural network systems for the prediction of discretized prosodic prominence. We show that pre-trained contextualized word representations from BERT outperform the other models even with less than 10% of the training data. Finally we discuss the dataset in light of the results and point to future research and plans for further improving both the dataset and methods of predicting prosodic prominence from text. The dataset and the code for the models are publicly available.
Original languageEnglish
Title of host publication22nd Nordic Conference on Computational Linguistics (NoDaLiDa) : Proceedings of the Conference
EditorsMareike Hartmann, Barbara Plank
Number of pages10
Place of PublicationLinköping
PublisherLinköping University Electronic Press
Publication date30 Sep 2019
ISBN (Electronic)978-91-7929-995-8
Publication statusPublished - 30 Sep 2019
MoE publication typeA4 Article in conference proceedings
EventNordic Conference on Computational Linguistics - Turku, Finland
Duration: 30 Sep 20192 Oct 2019
Conference number: 22

Publication series

NameLinköping Electronic Conference Proceedings
PublisherLinköping University Electronic Press
ISSN (Print)1650-3686
ISSN (Electronic)1650-3740
NameNEALT Proceedings Series

Fields of Science

  • 113 Computer and information sciences
  • Natural language processing
  • 6121 Languages

Cite this