The RAE and AWS use AI to know the state of Spanish on the Internet

The Royal Spanish Academy (RAE) y Amazon Web Services (AWS) have presented the spanish analysis tool that they have jointly created, with which they plan to examine tens of thousands of Internet documents at the same time in order to assess the state of Spanish in the world.

The tool has been developed from the AWS cloud-native technologies and with the advice of the RAE, and it is executed in three phases, which allows it to work with millions of documents, returning results in a very short time.

Currently tested with 8,745,563 digital texts of spontaneous Spanish from Spain and all Spanish-speaking countries in America, from social networks, forums or online sales platforms. A representation of journalistic texts has also been included to be able to observe the differences between one type of language and another.

With it, it seeks to assess the state of Spanish in the world, analyze the clarity of administrative language, compare the quality of Spanish by era or detect common errors in voice assistants and other AI devices. In its initial phase it allows identify foreign words, measure the lexical richness of a piece of writing and detect linguistic errors, as reported in a joint statement.

The collaboration between the RAE and AWS is part of the project Spanish Language and Artificial Intelligence (LEIA). The action seeks to apply AI to Spanish to analyze its current situation, take care of its use and ensure the unity of the language in all areas, especially digital.

The tool was presented this Thursday, at the headquarters of the Academy, in an act that was closed by the Secretary of State for Digitization and Artificial Intelligence, Carme Artigas, who has made a commitment to “Support the LEIA project with 5 million euros” from the Secretariat.

“The Spanish language applied to AI is going to be an unprecedented economic asset. We have 600 million Spanish speakers and we are determined to promote the development of this industry. In PERTE we have dedicated 100 million euros for the development of this corpus y 200 million more to promote an industry that develops AI applications in Spanish“, he added.

By Editor

Leave a Reply