mBART for Finnish text simplification
View resource name in all available languages
mBART suomenkielisen tekstin yksinkertaistamiseen
Persistent Identifier of this resource:
http://urn.fi/urn:nbn:fi:lb-2024011801
This resource will be available for download in Kielipankki – the Language Bank of Finland.
This is a neural model for Finnish sentence simplification. It is an mBART.CC25 (https://github.com/facebookresearch/fairseq/blob/main/examples/mbart/README.md) architecture finetuned on the texts included in the resource "Parallel Sentence Aligned Corpus of Finnish and Easy-to-read Finnish from the Yle News Archive 2014-2020, source" (http://urn.fi/urn:nbn:fi:lb-2024011703).
This model is provided as a binary file of 10GB. The model requires fairseq (https://github.com/facebookresearch/fairseq) to be installed along with Python.
People who looked at this resource also viewed the following:
- Parallel Sentence Aligned Corpus of Finnish and Easy-to-read Finnish from the Yle News Archive 2014-2020, source
- Parallel Corpus of Finnish and Easy-to-read Finnish from the Yle News Archive 2014-2018, source
- Parallel Corpus of Finnish and Easy-to-read Finnish from the Yle News Archive 2019-2020, source
- Christmas Gospel text-to-speech in four Uralic languages, source