UNMSM developed the first AI-generated news anchorwoman in Quechua

UNMSM developed the first AI-generated news anchorwoman in Quechua

Last Wednesday, the Faculty of Letters of the Universidad Nacional Mayor de San Marcos (UNMSM) announced that it had developed the first news narrator in Quechua generated by Artificial Intelligence (AI).

Illariywhich means “dawn” or “to break the dawn”, will be in charge of giving the most important news of the Dean of America each week, within the university newscast. Willakun TV Letters which is broadcasted on YouTube.

“Artificial intelligence is not the end product.”

In this regard, the professor Carlos Fernándezhead of Communications and Image of the UNMSM, gave details about the development of Illariy inside the classrooms of San Marino, as part of the production of the newsreel of the School of Letters.

“We said ‘we are going to renew this fifth season and we are going to include IA elements’ (…) We talked with the team (…) we were training for about a month, we did tests and, when they were positive, we began to project to Illariy“, he recounted in the program Family Dot Com.

“She had to be a certain age and, from there, we were thrown two possibilities. We fell in love with Illariy and we started to work with her (…) We had to think even about the outfit. We were very precise with the commands to get what we wanted,” he said.

Although the development of this level of artificial intelligence is a breakthrough in the field of technology and communications, the San Marino professor considered that Illariy still has many limitations.

Illariy is an accumulation of a series of processes, because artificial intelligence cannot be assembled without a human component. You have to use a series of programs to do it (…) There are talents that Illariy is not going to be able to develop because the intonation, the way we speak is still very far from it,” he said.

In this sense, he considered that artificial intelligence “is not the final product” but “the initial product” that allows us to start working.

“What ChatGPT can emit also contains. fake news(…) What we have to do is to start validating, analyzing the information, deepening it, debating it, because what cannot be done by the artificial intelligence is to generate critical thinking,” he said.

We recommend METADATARPP’s technology podcast. News, analysis, reviews, recommendations and everything you need to know about the technological world.


Daniel Chapman