Do machines speak Spanish?
Learn about how technology communicates and the role language plays in its development.
What language does technology communicate in?
To answer this question, we must also ask the following question: what does it mean for machines to speak in a certain language?
Infografía inteligencial artificial (2023) by Burhan KapkinerFundación Antonio de Nebrija
A language is much more than a set of sounds, grammatical rules, and a repertoire of words; speaking a language is based on a culture, heritage, and social practices.
And what does that mean?
Beliefs, biases, typical expressions, social norms, and the way of seeing the world translate into the internal configuration of different languages. Various factors have influenced the fact that English is currently the most widespread language in the field of AI.
The predominance of English in the field of machine communication and the importance of American centers in computer technology research could give rise to an American concept of communication.
Behind each machine, there is a human being who programs it and enters the necessary code for it to operate and interact.
If the language and culture of said person is American English, the machine will be highly influenced by these factors and it will show in the way it interacts, for example, in a chatbot.
Natural language: the new era of computing
Natural language is language spoken or written by humans for general communication purposes. It includes all characteristic expressions, traits, "imperfections," and cultural biases, as part of the DNA of language.
Secuencia del cortometraje Delirio (2021) by Carmen García RicoFundación Antonio de Nebrija
Language is our great weapon as human beings
For centuries, it has set us apart from all other living things on earth. Now, in the age of machines, they also have to "talk" (something we have achieved) and the challenge is making them speak in a similar way to humans.
This is where natural language processing, or NLP, technologies come in. Their purpose is to allow machines to understand language more deeply, to integrate into the culture that hosts the language, and to imitate the human way of speaking.
Language technology: language at the heart of AI
NLP research emerged in the Cold War, when the fear of espionage and an impending war prompted the first studies on machine translation aimed at quickly rendering texts from Russian to English.
Infografía inteligencial artificial (2023) by Burhan KapkinerFundación Antonio de Nebrija
What's new?
The way in which we program artificial intelligence: machine learning and deep learning, rely on the functioning of the human brain and its neural connections to reproduce them in the form of AI.
This is in addition to language technology or LT, which studies a language in depth in order to integrate it into the processing of machines. There are already many examples of LT that we use on a daily basis, such as voice recognition or automatic subtitles.
Will machines speak Spanish?
Spanish is the second most common mother tongue in the world and the fourth most spoken language. It is expected that by 2030 it will be the second language in international trade, due to the economic growth of Latin America and the impact of this language on the North American market and society.
Infografía inteligencial artificialFundación Antonio de Nebrija
It is essential that machines speak and understand Spanish and the cultures that surrounds it. For this reason, different projects have been undertaken in Spain aimed at promoting the development of language technologies.
Spain is a pioneer in this area with the creation of the Spanish Society for Natural Language Processing (SEPLN) in 1983, whose objective is the promotion, research, development, and teaching of NLP in Spanish at a national and international level.
The support of internationally renowned institutions, such as the National Library of Spain, the RAE, and the academies of ASALE, the Cervantes Institute, or RTVE, is important in the quest to achieve the development of AI capable of understanding Spanish in its entirety.
We've consulted the following study for this article: Cómo hablan las máquinas español: logros, retos y oportunidades para a inteligencia artificial aplicada al lenguaje.
Content curation: Aina Arbona
Graphic Editor: Carmen García
Consulting and review: Lola Pons (Universidad de Sevilla)
