Abstract : We address the design of a unified multilingual system for handwriting recognition. Most of multilingual systems rests on specialized models that are trained on a single language and one of them is selected at test time. While some recognition systems are based on a unified optical model, dealing with a unified language model remains a major issue, as traditional language models are generally trained on corpora composed of large word lexicons per language. Here, we bring a solution by considering language models based on sub-lexical units, called multigrams. Dealing with multigrams strongly reduces the lexicon size and thus decreases the language model complexity. This makes possible the design of an end-to-end unified multilingual recognition system where both a single optical model and a single language model are trained on all the languages. We discuss the impact of the language unification on each model and show that our system reaches state-of-the-art methods performance with a strong reduction of the complexity.
https://hal-normandie-univ.archives-ouvertes.fr/hal-02075654 Contributor : Accord Elsevier CCSDConnect in order to contact the contributor Submitted on : Friday, October 22, 2021 - 11:27:54 AM Last modification on : Wednesday, March 2, 2022 - 10:10:12 AM Long-term archiving on: : Sunday, January 23, 2022 - 7:18:54 PM