L’équipe GETALP crée ou contribue à la création de ressources de plusieurs types (corpus, ressources lexicales, logiciels, etc.) qui sont habituellement mises à disposition du public et des autres chercheurs. Vous trouverez ici un extrait des ressources qui ont été créé par le GETALP ou auxquelles le GETALP a contribué.
Resources
| Title | Dates | Licence | Availability | Purpose/Summary |
|---|---|---|---|---|
| Audiocite.net | 2024 | CC BY-NC-SA 4.0 | https://openslr.org/139/ | Audiocite.net is a corpus of read French speech downloaded in November 2021 from the Audiocite.net website. |
| TArC: Tunisian Arabish Corpus | 2024 (2021-) | CC BY-NC-SA 4.0 | https://github.com/eligugliotta/tarc | A semi-automatically annotated corpus of Tunisian Arabic, including classification of tokens (Tunisian Arabic, foreign language, emotag), transcription into Arabic script, tokenization, and morpho-syntactic tagging. |
| LeBenchmark | 2024 (2022-) | Apache-2.0 | https://huggingface.co/LeBenchmark | A reproducible and multifaceted benchmark for evaluating speech SSL models, focusing on four French language tasks: Speech Recognition (ASR), Spoken Language Understanding (SLU), Speech Translation (AST), and Emotion Recognition (AER). More information on the project page http://lebenchmark.com |
| PxCorpus | 2023 (2022-) | Creative Commons Attribution 4.0 International | DOI 10.5281/zenodo.6482586 | PxCorpus (A Spoken Drug Prescription Dataset in French for Spoken Language Understanding and Dialogue) is the first distributed spoken medical drug prescription corpus in French. It contains 4 hours of transcribed and annotated dialogues from 55 participants (experts and non-experts) and is designed for training NLP models for Spoken Language Understanding and dialogue systems. Transcriptions were human-verified and aligned with semantic labels. With was a collaboration with CHU Grenoble and Calystène SA. |
| DBnary | 2025 (2012-) | Creative Commons Attribution-ShareAlike 3.0 | http://kaiko.getalp.org/ | Dbnary is an effort to provide multilingual lexical data extracted from Wiktionary. The extracted data is made available as LLOD (Linguistic Linked Open Data). This data set won the Monnet challenge in 2012. Since then a new version is issued twice a month and new languages and data is added continuously. |
| BDBRUIT | 2005 | ELRA | https://catalogue.elra.info/en-us/repository/browse/ELRA-S0033/ | A French speech database dedicated to the study of the perturbations of speech production due to noisy environments, and especially the Lombard effect. Environment: 4 noise conditions and the reference condition (quiet). The 2 noises used (a "white noise" and a "cocktail-party noise") were both produced to the ears of the speaker first without then with the feedback of his own voice. 10 speakers: 5 male and 5 female (4 CDROMs, approximately 2.2 Gigabytes) |
| BDSONS | 2005 | ELRA | https://catalogue.elra.info/en-us/repository/browse/ELRA-S0005/ | The BDSONS Database is a French - speech database with two subsets: evaluation and acoustic modelling. The Corpora consist of 32 speakers: 16 male and 16 female (7 CD-ROMs of approximately 3,5 Gigabytes) |
| BRAF100 | 2005 | ELRA | https://catalogue.elra.info/en-us/repository/browse/ELRA-S0197/ | GETALP (GEOD at that time) recorded a large oral database (BRAF100: Base pour la Reconnaissance Automatique du Français with 100 speakers - around 30h of speech) within a cooperation contract with ISL (Interactive Systems Laboratories) at the University of Karlsruhe for the joint development of a French speech database. This corpus has been distributed as part of GlobalPhone corpus on ELRA |
| THERADIA WoZ | 2025 | End User License Agreement (EULA) | https://catalog.elra.info/ (summer 2025) | Ecological corpus specifically designed for the audiovisual detection of affective states in the context of healthcare domain. The corpus data come from natural interactions with a virtual assistant, operated remotely by a human acting as a Wizard-of-Oz, in French, involving 61 senior individuals, including both neurotypical and Mild Cognitive Impairment (MCI) participants, and 52 youth individuals. The participants’ expressions were fully transcribed and partially annotated based on the dimensions of recent appraisal theories models and 23 affective labels derived from the literature on achievement affects. |
| Vlexique 2.0 | 2024 | Creative Commons Attribution Non Commercial Share Alike 4.0 International | https://doi.org/10.5281/zenodo.14069226 | Vlexique 2.0 is a collection of French verbal paradigms, in phonemic and orthographic notation. They are suited for both computational and manual analysis. The dataset follows the Paralex Standard (https://www.paralex-standard.org). |
Affichage des éléments 1 à 10 sur 40 éléments
Plus
Pointeurs vers des collections de ressources créées en tout ou partie par le GETALP :
collection GETALP sur Zenodo
groupe GETALP sur gricad-gitlab
groupe GETALP sur github