Rechercher

We present lemmatization experiments on the unstandardized low-resourced languages Low Saxon and Occitan using two machine-learningbased approaches represented by MaChAmp and Stanza. We show different ways to increase training data by leveraging historical corpora, small amounts of gold data and dictionary information, and discuss the usefulness of this additional data. In the results, we find some differences in the performance of the models depending on the language. This variation is likely to be partly due to differences in the corpora we used, such as the amount of internal variation. However, we also observe common tendencies, for instance that sequential models trained only on gold-annotated data often yield the best overall performance and generalize better to unknown tokens.

We study the capabilities of speech processing systems trained simply to predict large amounts of transcripts of audio on the internet. When scaled to 680,000 hours of multilingual and multitask supervision, the resulting models generalize well to standard benchmarks and are often competitive with prior fully supervised results without the need for any dataset specific fine-tuning. When compared to humans, the models approach their accuracy and robustness. We are releasing models and inference code to serve as a foundation for further work on robust speech processing.

Dans l’esprit d’une majorité de Français, les langues dites régionales ne seraient que des « patois », de vulgaires déformations du français, de vagues idiomes tout juste bons à décrire des banalités. Pourquoi devraient-ils s’émouvoir de leur effacement ? Or, tous les linguistes le savent : le basque, le breton, l’alsacien, le corse, le picard et les autres, n’ont rien à envier au français, à l’anglais, à l’arabe ou au mandarin. La seule différence entre les « petites langues » et les autres, c’est que les premières n’ont pas eu la chance de devenir des langues officielles d’un État. Cet ouvrage affiche une ambition assumée : réconcilier la France avec sa diversité. Pour que le français reste notre langue commune, sans devenir notre langue unique.

This paper presents OcWikiDisc, a new freely available corpus in Occitan, as well as language identification experiments on Occitan done as part of the corpus building process. Occitan is a regional language spoken mainly in the south of France and in parts of Spain and Italy. It exhibits rich diatopic variation, it is not standardized, and it is still low-resourced, especially when it comes to large downloadable corpora. We introduce OcWikiDisc, a corpus extracted from the talk pages associated with the Occitan Wikipedia. The version of the corpus with the most restrictive language filtering contains 8K user messages for a total of 618K tokens. The language filtering is performed based on language identification experiments with five off-the-shelf tools, including the new fasttext's language identification model from Meta AI's No Language Left Behind initiative, released in July 2022.

We investigate methods to develop a parser for Martinican Creole, a highly under-resourced language, using a French treebank. We compare transfer learning and multi-task learning models and examine different input features and strategies to handle the massive size imbalance between the treebanks. Surprisingly, we find that a simple concatenated (French + Martinican Creole) baseline yields optimal results even though it has access to only 80 Martinican Creole sentences. POS embeddings work better than lexical ones, but they suffer from negative transfer.

Driven by the goal of eradicating language barriers on a global scale, machine translation has solidified itself as a key focus of artificial intelligence research today. However, such efforts have coalesced around a small subset of languages, leaving behind the vast majority of mostly low-resource languages. What does it take to break the 200 language barrier while ensuring safe, high quality results, all while keeping ethical considerations in mind? In No Language Left Behind, we took on this challenge by first contextualizing the need for low-resource language translation support through exploratory interviews with native speakers. Then, we created datasets and models aimed at narrowing the performance gap between low and high-resource languages. More specifically, we developed a conditional compute model based on Sparsely Gated Mixture of Experts that is trained on data obtained with novel and effective data mining techniques tailored for low-resource languages. We propose multiple architectural and training improvements to counteract overfitting while training on thousands of tasks. Critically, we evaluated the performance of over 40,000 different translation directions using a human-translated benchmark, Flores-200, and combined human evaluation with a novel toxicity benchmark covering all languages in Flores-200 to assess translation safety. Our model achieves an improvement of 44% BLEU relative to the previous state-of-the-art, laying important groundwork towards realizing a universal translation system. Finally, we open source all contributions described in this work, accessible at https://github.com/facebookresearch/fairseq/tree/nllb.

This paper describes a method of semi-automatic word spotting in minority languages, from one and the same Aesop fable “The North Wind and the Sun” translated in Romance languages/dialects from Hexagonal (i.e. Metropolitan) France and languages from French Polynesia. The first task consisted of finding out how a dozen words such as “wind” and “sun” were translated in over 200 versions collected in the field — taking advantage of orthographic similarity, word position and context. Occurrences of the translations were then extracted from the phone-aligned recordings. The results were judged accurate in 96–97% of cases, both on the development corpus and a test set of unseen data. Corrected alignments were then mapped and basemaps were drawn to make various linguistic phenomena immediately visible. The paper exemplifies how regular expressions may be used for this purpose. The final result, which takes the form of an online speaking atlas (enriching the https://atlas.limsi.fr website), enables us to illustrate lexical, morphological or phonetic variation.

Machine translation has been researched using deep neural networks in recent years. These networks require lots of data to learn abstract representations of the input stored in continuous vectors. Dialect translation has become more important since the advent of social media. In particular, when dialect speakers and standard language speakers no longer understand each other, machine translation is of rising concern. Usually, dialect translation is a typical low-resourced language setting facing data scarcity problems. Additionally, spelling inconsistencies due to varying pronunciations and the lack of spelling rules complicate translation. This paper presents the best-performing approaches to handle these problems for Alemannic dialects. The results show that back-translation and conditioning on dialectal manifestations achieve the most remarkable enhancement over the baseline. Using back-translation, a significant gain of +4.5 over the strong transformer baseline of 37.3 BLEU points is accomplished. Differentiating between several Alemannic dialects instead of treating Alemannic as one dialect leads to substantial improvements: Multi-dialectal translation surpasses the baseline on the dialectal test sets. However, training individual models outperforms the multi-dialectal approach. There, improvements range from 7.5 to 10.6 BLEU points over the baseline depending on the dialect.

Identifying phone inventories is a crucial component in language documentation and the preservation of endangered languages. However, even the largest collection of phone inventory only covers about 2000 languages, which is only 1/4 of the total number of languages in the world. A majority of the remaining languages are endangered. In this work, we attempt to solve this problem by estimating the phone inventory for any language listed in Glottolog, which contains phylogenetic information regarding 8000 languages. In particular, we propose one probabilistic model and one non-probabilistic model, both using phylogenetic trees (“language family trees”) to measure the distance between languages. We show that our best model outperforms baseline models by 6.5 F1. Furthermore, we demonstrate that, with the proposed inventories, the phone recognition model can be customized for every language in the set, which improved the PER (phone error rate) in phone recognition by 25%.

Poète, pédagogue, lexicographe, fabuliste et traducteur, Hector Poullet est un acteur majeur de la scène locale et un incorrigible polygraphe. A un âge où d’autres se retirent, il se voit aujourd’hui encore sollicité pour enseigner la langue ou pour assurer des conférences savantes. L’examen de sa bibliographie accuse un compte de trente-cinq ouvrages, publiés seul ou en équipe, dont on trouve ici un relevé. A cette production respectable, il faut ajouter nombre de préfaces, de traductions, de chapitres de livres, d’articles inclus dans des sommes collectives ou sur divers sites. Suivant la proposition de la Direction des Affaires Culturelles, nous avons conçu un film d’entretien, où l’on entend cette personnalité évoquer les épisodes significatifs de l’histoire récente de la Guadeloupe. La présente contribution ne pouvait se concevoir ni comme le verbatim de ce dialogue, ni comme le simple script du film. Nous choisissons donc d’éclairer deux épisodes mal connus de son action dans l’entrée du créole au collège et dans l’élaboration du premier dictionnaire du guadeloupéen. Itinérance et diversité Né en 1938, Hector aime rappeler la diversité de ses origines et le nomadisme de son enfance, avec un père originaire de Grande Terre et une mère de Basse Terre. Il vit une enfance de fils d’instituteur itinérant, balloté au gré des affectations parentales, aux quatre coins de l’archipel. Le bac en poche, il va à Paris pour poursuivre des études de sciences. Au fil d’un séjour qui s

Grapheme-to-Phoneme (G2P) has many applications in NLP and speech fields. Most existing work focuses heavily on languages with abundant training datasets, which limits the scope of target languages to less than 100 languages. This work attempts to apply zero-shot learning to approximate G2P models for all low-resource and endangered languages in Glottolog (about 8k languages). For any unseen target language, we first build the phylogenetic tree (i.e. language family tree) to identify top-k nearest languages for which we have training sets. Then we run models of those languages to obtain a hypothesis set, which we combine into a confusion network to propose a most likely hypothesis as an approximation to the target language. We test our approach on over 600 unseen languages and demonstrate it significantly outperforms baselines.

Languages are classified as low-resource when they lack the quantity of data necessary for training statistical and machine learning tools and models. Causes of resource scarcity vary but can include poor access to technology for developing these resources, a relatively small population of speakers, or a lack of urgency for collecting such resources in bilingual populations where the second language is high-resource. As a result, the languages described as low-resource in the literature are as different as Finnish on the one hand, with millions of speakers using it in every imaginable domain, and Seneca, with only a small-handful of fluent speakers using the language primarily in a restricted domain. While issues stemming from the lack of resources necessary to train models unite this disparate group of languages, many other issues cut across the divide between widely-spoken low-resource languages and endangered languages. In this position paper, we discuss the unique technological, cultural, practical, and ethical challenges that researchers and indigenous speech community members face when working together to develop language technology to support endangered language documentation and revitalization. We report the perspectives of language teachers, Master Speakers and elders from indigenous communities, as well as the point of view of academics. We describe an ongoing fruitful collaboration and make recommendations for future partnerships between academic researchers and language community stakeholders.

Apertium linguistic data for Occitan

L’Alsace peut être qualifiée de province au même titre que la Bourgogne ou la Franche-Comté, par exemple ; il y aura peu de désaccord à propos d’une telle dénomination. Mais quand il s’agit de définir la situation linguistique de l’Alsace, des difficultés nombreuses surgissent qui sont liées au fait qu’il ne s’agit pas seulement de repérer des langues en usage par rapport à une classification des langues fondée sur des critères scientifiques mais qu’il s’agit aussi de les situer les unes par ...