Code-switched inspired losses for generic spoken dialog representations - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing Année : 2021

Code-switched inspired losses for generic spoken dialog representations

Résumé

Spoken dialog systems need to be able to handle both multiple languages and multilinguality inside a conversation (e.g in case of codeswitching). In this work, we introduce new pretraining losses tailored to learn multilingual spoken dialog representations. The goal of these losses is to expose the model to codeswitched language. To scale up training, we automatically build a pretraining corpus composed of multilingual conversations in five different languages (French, Italian, English, German and Spanish) from OpenSubtitles, a huge multilingual corpus composed of 24.3G tokens. We test the generic representations on MIAM, a new benchmark composed of five dialog act corpora on the same aforementioned languages as well as on two novel multilingual downstream tasks (i.e multilingual mask utterance retrieval and multilingual inconsistency identification). Our experiments show that our new code switched-inspired losses achieve a better performance in both monolingual and multilingual settings.
Fichier principal
Vignette du fichier
2108.12465.pdf (624.56 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03574595 , version 1 (15-02-2022)

Identifiants

  • HAL Id : hal-03574595 , version 1

Citer

Emile Chapuis, Pierre Colombo, Matthieu Labeau, Chloé Clavel. Code-switched inspired losses for generic spoken dialog representations. 2021 Conference on Empirical Methods in Natural Language Processing, Nov 2021, Punta Cana, Dominican Republic. ⟨hal-03574595⟩
34 Consultations
42 Téléchargements

Partager

Gmail Facebook X LinkedIn More