The fields of artificial intelligence (AI) and natural language processing (NLP) are rapidly advancing — recent developments in these fields allow for individuals to translate human language into computer code or develop highly efficient speech recognition systems for under-resourced languages. But when it comes to signed languages like American Sign Language (ASL), NLP research often ignores these types of languages.

Modern AI technologies are capable of producing highly complex written language, yet there’s a clear disparity in terms of the amount of NLP research on spoken languages versus signed languages. In a recently published paper entitled “Including Signed Languages in Natural Language Processing,” an international team of NLP researchers explored the importance of addressing this lack of NLP research on signed languages.

“Signed languages, even though they are a significant part of the languages used in the world, aren’t included (in most NLP research),” said Kayo Yin, one of the paper’s co-authors. “There is a demand and an importance in having technology that can handle signed languages.”

The paper was presented at the 59th Annual Meeting of the Association for Computational Linguistics, where it won the Best Theme Paper award. In the paper, the researchers argue that while NLP research focusing on spoken languages is abundant, speakers of signed languages are not able to fully take advantage of such technologies.

“In a predominantly oral society, deaf people are constantly encouraged to use spoken languages through lip reading or text-based communication” the paper reads. “The exclusion of signed languages from modern language technologies further suppresses signing in favor of spoken languages.”

According to the paper, there has been a decent amount of research on signed languages in another, adjacent scientific field: computer vision. This field focuses on understanding the ways in which computer programs can analyze and interpret information from videos and digital images — however, computer vision research rarely deals with the different aspects of human language, such as syntax and semantics. The researchers end the paper with a call for researchers in NLP and computer vision to coordinate with each other in order to create better technologies — for instance, machine translation systems — for users of signed languages.

“We hope to see an increase in both the interests and efforts in collecting signed language resources and developing signed language tools while building a strong collaboration with signing communities,” the paper concludes.