Hierarchical Pre-training for Sequence Labelling in Spoken Dialog

Emile Chapuis, Pierre Colombo, Matteo Manica, Matthieu Labeau, Chloé Clavel


Abstract
Sequence labelling tasks like Dialog Act and Emotion/Sentiment identification are a key component of spoken dialog systems. In this work, we propose a new approach to learn generic representations adapted to spoken dialog, which we evaluate on a new benchmark we call Sequence labellIng evaLuatIon benChmark fOr spoken laNguagE benchmark (SILICONE). SILICONE is model-agnostic and contains 10 different datasets of various sizes. We obtain our representations with a hierarchical encoder based on transformer architectures, for which we extend two well-known pre-training objectives. Pre-training is performed on OpenSubtitles: a large corpus of spoken dialog containing over 2.3 billion of tokens. We demonstrate how hierarchical encoders achieve competitive results with consistently fewer parameters compared to state-of-the-art models and we show their importance for both pre-training and fine-tuning.
Anthology ID:
2020.findings-emnlp.239
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2020
Month:
November
Year:
2020
Address:
Online
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
2636–2648
Language:
URL:
https://aclanthology.org/2020.findings-emnlp.239
DOI:
10.18653/v1/2020.findings-emnlp.239
Bibkey:
Cite (ACL):
Emile Chapuis, Pierre Colombo, Matteo Manica, Matthieu Labeau, and Chloé Clavel. 2020. Hierarchical Pre-training for Sequence Labelling in Spoken Dialog. In Findings of the Association for Computational Linguistics: EMNLP 2020, pages 2636–2648, Online. Association for Computational Linguistics.
Cite (Informal):
Hierarchical Pre-training for Sequence Labelling in Spoken Dialog (Chapuis et al., Findings 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.findings-emnlp.239.pdf
Optional supplementary material:
 2020.findings-emnlp.239.OptionalSupplementaryMaterial.pdf
Data
SILICONE BenchmarkDailyDialogEmotionLinesIEMOCAPMELDMRDAOpenSubtitlesSEMAINESwitchboard-1 Corpus