Neural text generation is receiving broad attention with the publication of new tools such as ChatGPT. The main reason for that is that the achieved quality of the generated text may be attributed to a human writer by the naked eye of a human evaluator. In this paper, we propose a new corpus in French and English for the task of recognising automatically generated texts and we conduct a study of how humans perceive the text. Our results show, as previous work before the ChatGPT era, that the generated texts by tools such as ChatGPT share some common characteristics but they are not clearly identifiable which generates different perceptions of these texts.

Intent classification and slot-filling are essential tasks of Spoken Language Understanding (SLU). In most SLU systems, those tasks are realized by independent modules, but for about fifteen years, models achieving both of them jointly and exploiting their mutual enhancement have been proposed. A multilingual module using a joint model was envisioned to create a touristic dialogue system for a European project, HumanE-AI-Net. A combination of multiple datasets, including the MEDIA dataset, was suggested for training this joint model. The MEDIA SLU dataset is a French dataset distributed since 2005 by ELRA, mainly used by the French research community and free for academic research since 2020. Unfortunately, it is annotated only in slots but not intents. An enhanced version of MEDIA annotated with intents has been built to extend its use to more tasks and use cases. This paper presents the semi-automatic methodology used to obtain this enhanced version. In addition, we present the first results of SLU experiments on this enhanced dataset using joint models for intent classification and slot-filling.

Les tâches de détection d’intention et d’identification des concepts sont toutes deux des éléments importants de la compréhension de la parole. Elles sont souvent réalisées par deux modules différents au sein d’un pipeline. L’apparition de modèles réalisant conjointement ces deux tâches a permis d’exploiter les dépendances entre elles et d’améliorer les performances obtenues. Plus récemment, des modèles de détection jointe reposant sur des architectures Transformer ont été décrits dans la littérature. Par ailleurs, avec la popularité et taille croissante des modèles Transformer ainsi que les inquiétudes ergonomiques et écologiques grandissantes, des modèles compacts ont été proposés. Dans cet article, nous présentons la mise en place et l’évaluation d’un modèle compact pour la détection jointe de l’intention et des concepts. Notre contexte applicatif est celui d’un système interactif de questions-réponses français.