Findings of the AmericasNLP 2021 Shared Task on Open Machine Translation for Indigenous Languages of the Americas

Mager, Manuel; Oncevay, Arturo; Ebrahimi, Abteen; Ortega, John E.; Rios, Annette; Fan, Angela; Gutierrez-Vasques, Ximena; Chiruzzo, Luis; Gimenez-Lugo, Gustavo A.; Ramos, Ricardo; Meza, Iván; Coto-Solano, Rolando; Palmer, Alexis; Mager-Hois, Elisabeth; Chaudhary, Vishrav; Neubig, Graham; Vu, Ngoc Thang; Kann, Katharina

doi:10.18653/v1/2021.americasnlp-1.23

“…Our 2020), the numerous results on the largely South-and Central American languages from the first AmericasNLP conference (Mager et al, 2021), and the Inuktitut language of Canada (Joanis et al, 2020). Datasets for lower-resourced languages of India have also sprung up, including the 13-language PMIndia , and datasets focused on languages of the Northeast like Mizo (Thihlum et al, 2020), Khasi (Laskar et al, 2021) and Assamese (Laskar et al, 2020).…”

Section: Related Workmentioning

confidence: 86%

“…like AfricaNLP 6 , AmericasNLP(Mager et al, 2021) and ArabNLP 7 . These efforts, in addition to providing datasets, frequently provide models and baselines, or even public interfaces, like the Khaya Translator Web App 8 by GhanaNLP for West African languages, and the lesan.ai 9 translation website for Ethiopian languages.…”

mentioning

confidence: 99%

Building Machine Translation Systems for the Next Thousand Languages

Bapna¹,

Caswell²,

Kreutzer³

et al. 2022

Preprint

View full text Add to dashboard Cite

In this paper we share findings from our effort to build practical machine translation (MT) systems capable of translating across over one thousand languages. We describe results in three research domains: (i) Building clean, web-mined datasets for 1500+ languages by leveraging semi-supervised pre-training for language identification and developing data-driven filtering techniques; (ii) Developing practical MT models for under-served languages by leveraging massively multilingual models trained with supervised parallel data for over 100 high-resource languages and monolingual datasets for an additional 1000+ languages; and (iii) Studying the limitations of evaluation metrics for these languages and conducting qualitative analysis of the outputs from our MT models, highlighting several frequent error modes of these types of models. Using this approach, we add 24 new languages to Google Translate, the product's largest increase in language coverage to-date. We hope that our work provides useful insights to practitioners working towards building MT systems for currently understudied languages, and highlights research directions that can complement the weaknesses of massively multilingual models in data-sparse settings.

show abstract

“…The fourth, fifth, and sixth contributions are all related to the first-time presentation of a deep learning transformer model for Quechua that is used for NER and POS classification tasks. One of the works that presented deep learning approaches for Quechua is a shared task (Mager et al, 2021a) from the first workshop on NLP for indigenous languages of the America (Mager et al, 2021b). Another work called indt5 (Nagoudi et al, 2021) used an encoder-decoder model transformers based on T5 (Raffel et al, 2020).…”

Section: Related Workmentioning

confidence: 99%

Introducing QuBERT: A Large Monolingual Corpus and BERT Model for Southern Quechua

Zevallos¹,

Ortega²,

Chen³

et al. 2022

Proceedings of the Third Workshop on Deep Learning for Low-Resource Natural Language Processing

Self Cite

2

0

View full text Add to dashboard Cite

The lack of resources for languages in the Americas has proven to be a problem for the creation of digital systems such as machine translation, search engines, chat bots, and more. The scarceness of digital resources for a language causes a higher impact on populations where the language is spoken by millions of people. We introduce the first official large combined corpus for deep learning of an indigenous South American low-resource language spoken by millions called Quechua. Specifically, our curated corpus is created from text gathered from the southern region of Peru where a dialect of Quechua is spoken that has not traditionally been used for digital systems as a target dialect in the past. In order to make our work repeatable by others, we also offer a public, pre-trained, BERT model called Qu-BERT which is the largest linguistic model ever trained for any Quechua type, not just the southern region dialect. We furthermore test our corpus and its corresponding BERT model on two major tasks: (1) named-entity recognition (NER) and (2) part-of-speech (POS) tagging by using state-of-the-art techniques where we achieve results comparable to other work on higher-resource languages. In this article, we describe the methodology, challenges, and results from the creation of QuBERT which is on on par with other state-of-the-art multilingual models for natural language processing achieving between 71 and 74% F1 score on NER and 84-87% on POS tasks.

show abstract

“…The fourth, fifth, and sixth contributions are all related to the first-time presentation of a deep learning transformer model for Quechua that is used for NER and POS classification tasks. One of the works that presented deep learning approaches for Quechua is a shared task (Mager et al, 2021a) from the first workshop on NLP for indigenous languages of the America (Mager et al, 2021b). Another work called indt5 (Nagoudi et al, 2021) used an encoder-decoder model transformers based on T5 .…”

Section: Related Workmentioning

confidence: 99%

Proceedings of the Third Workshop on Deep Learning for Low-Resource Natural Language Processing

2022

0

View full text Add to dashboard Cite

The lack of resources for languages in the Americas has proven to be a problem for the creation of digital systems such as machine translation, search engines, chat bots, and more. The scarceness of digital resources for a language causes a higher impact on populations where the language is spoken by millions of people. We introduce the first official large combined corpus for deep learning of an indigenous South American low-resource language spoken by millions called Quechua. Specifically, our curated corpus is created from text gathered from the southern region of Peru where a dialect of Quechua is spoken that has not traditionally been used for digital systems as a target dialect in the past. In order to make our work repeatable by others, we also offer a public, pre-trained, BERT model called Qu-BERT which is the largest linguistic model ever trained for any Quechua type, not just the southern region dialect. We furthermore test our corpus and its corresponding BERT model on two major tasks: (1) named-entity recognition (NER) and (2) part-of-speech (POS) tagging by using state-of-the-art techniques where we achieve results comparable to other work on higher-resource languages. In this article, we describe the methodology, challenges, and results from the creation of QuBERT which is on on par with other state-of-the-art multilingual models for natural language processing achieving between 71 and 74% F1 score on NER and 84-87% on POS tasks. ReferencesWillem FH Adelaar. 2004. The languages of the Andes.

show abstract

Findings of the AmericasNLP 2021 Shared Task on Open Machine Translation for Indigenous Languages of the Americas

Cited by 25 publications

References 25 publications

Building Machine Translation Systems for the Next Thousand Languages

Building Machine Translation Systems for the Next Thousand Languages

Introducing QuBERT: A Large Monolingual Corpus and BERT Model for Southern Quechua

Proceedings of the Third Workshop on Deep Learning for Low-Resource Natural Language Processing

Contact Info

Product

Resources

About