This is the program capable of translating indigenous Mexican languages ​​like the Google translator

https://mundo.sputniknews.com/20211231/the-program-capable-of-translating-indigenous-mexican-languages-like-the-google-translator-1119927106.html
This is the program capable of translating indigenous Mexican languages ​​like the Google translator
This is the program capable of translating indigenous Mexican languages ​​like the Google translator
Researchers from the National Autonomous University of Mexico (UNAM) develop a program that will automatically translate indigenous languages ​​such as Huichol... 12.31.2021, Sputnik Mundo
2021-12-31T21:14+0000
2021-12-31T21:14+0000
2021-12-31T21:40+0000
Google
National Autonomous University of Mexico (UNAM)
indigenous languages
Microsoft
Mexico
technology
/html/head/meta[@name='og:title']/@content
/html/head/meta[@name='og:description']/@content
https://cdnnmundo1.img.sputniknews.com/img/07e5/0b/11/1118335989_0:0:1280:721_1920x0_80_0_0_6490f071383889af8d4df74a27174103.jpg
The project is developed by Iván Vladimir Meza, a researcher at the Department of Computer Science at the Institute for Applied Mathematics and Systems Research (IIMAS) at UNAM and an expert in artificial intelligence. The specialist initiated the project of a translator of indigenous languages ​​in 2014, thanks to the fact that one of his students had a relationship with the Wixárika community (known as Huicholes) and wanted to support it. To process all the information, Meza Ruiz will use the computational model of neural networks, a process that, as its name indicates, resembles the way in which the brain processes a language.Basically what this model does is to do the translation based on examples and comparison of sentences, also relying on common mathematical concepts such as matrix operations and vector calculus."Fortunately we have several algorithms that work fine, but since current so-called deep models have numerous modules and values ​​for p To process, specialized computer equipment is needed," explained Meza Ruiz in an interview for the General Directorate of Social Communication of the UNAM. Unfortunately, one of the biggest challenges of the project is the little written information that is available on the 68 indigenous languages recognized in Mexico (with 364 variants), since even many survive only orally." For native languages, the largest corpus are close to 10,000 examples, compared to the millions of commercial systems. We are very far from having an experience similar to what we have when using a normal translator, because we have very little data. That is part of our battle right now: to get more data and increase our examples," explained Delfino Zacarias, a master's student who speaks Ayuuk (Mixe), who collaborates with an interpretation method. Volunteers such as César Cruz, undergraduate student at IIMAS, who documents an intelligent system for mazahua (J natío) and which he developed in the MazhuApp app, available in the Play Store. Other universities such as Querétaro and Yucatán have developed software in collaboration with Microsoft to translate Otomi and Mayan in a program called Heritage.
https://mundo.sputniknews.com/20211221/campesinas-e-indigenas-de-america-latina-search-to-be-owners-of-the-land-they-work-1119579349.html
https://mundo.sputniknews.com/20211220/indigenas-se-opponen-al-amlo-agreement-on-mega-works-1119532070.html
Mexico
Sputnik World
contact@sputniknews.com
+74956456601
MIA „Rosiya Segodnya“
252
60
2021
Sputnik World
contact@sputniknews.com
+74956456601
MIA „Rosiya Segodnya“
252
60
News
en_ES
Sputnik World
contact@sputniknews.com
+74956456601
MIA „Rosiya Segodnya“
252
60
1920
1080
true
1920
1440
true
https://cdnnmundo1.img.sputniknews.com/img/07e5/0b/11/1118335989_103:0:1206:827_1920x0_80_0_0_6faeed515be32aee004ef0c260eadbc8.jpg
1920
1920
true
Sputnik World
contact@sputniknews.com
+74956456601
MIA „Rosiya Segodnya“
252
60
Sputnik World
contact@sputniknews.com
+74956456601
MIA „Rosiya Segodnya“
252
60
google, national autonomous university of mexico (unam), indigenous languages, microsoft, mexico
21:14 GMT 12.31.2021 (updated: 21:40 GMT 12.31.2021)
© Photo : Pixabay / PublicDomainPicturesDictionary (reference image)
© Photo : Pixabay / PublicDomainPictures
Follow us on
Researchers from the National Autonomous University of Mexico (UNAM) are developing a program that will automatically translate indigenous languages ​​such as Huichol, Nahuatl, Mexicanero and Yoem Noli and Mazahua, as well as translators such as Google.
The project is developed by Iván Vladimir Meza, a researcher at the Department of Computer Science at the Institute for Research in Applied Mathematics and Systems (IIMAS) at UNAM and an expert in artificial intelligence.
The specialist started the project of a translator of indigenous languages ​​in 2014, thanks to the fact that one of his students had a relationship with the Wixárika community (known as Huicholes) and wanted to support it.
Latin America
Peasant and indigenous women of Latin America seek to own the land they work
December 21, 2021, 19:05 GMT
To process all the information, Meza Ruiz will use the computational model of neural networks, a process that, as its name indicates, resembles the way in which the brain processes a language.
Basically what this model does is make the translation based on examples and comparison of sentences, also relying on common mathematical concepts such as matrix operations and vector calculus.
"Fortunately we have several algorithms that work well, but since the current so-called deep models have numerous modules and values ​​to process, specialized computer equipment is needed," Meza Ruiz explained in an interview for the General Directorate of Social Communication of the UNAM.
Latin America
Indigenous people oppose AMLO's agreement on mega-projects
December 20, 2021, 18:32 GMT
Unfortunately, one of the biggest challenges of the project is the little written information that is available on the 68 indigenous languages ​​recognized in Mexico (with 364 variants), since even many survive only orally.
"For native languages, the largest corpora are close to 10,000 examples, compared to millions for commercial systems. We are very far from having an experience similar to what we have when using a normal translator, because we have very little data. That it is part of our battle right now: to get more data and increase our examples," explained Delfino Zacarias, a master's student, Ayuuk (Mixe) speaker, who collaborates with an interpretation method.
Volunteers have joined the project such as César Cruz, an undergraduate student at IIMAS, who documents an intelligent system for Mazahua (J natío) and which he developed in the MazhuApp app, available in the Play Store.
Other universities such as Querétaro and Yucatán have developed in collaboration with Microsoft software to translate Otomí and Mayan in a program called Heritage.

Este es el programa capaz de traducir lenguas indígenas mexicanas como el traductor de Google