Mixtec–Spanish Parallel Text Dataset for Language Technology Development

This article introduces a freely available Spanish–Mixtec parallel corpus designed to foster natural language processing (NLP) development for an indigenous language that remains digitally low-resourced. The dataset, comprising 14,587 sentence pairs, covers Mixtec variants from Guerrero (Tlacoachist...

Full description

Saved in:
Bibliographic Details
Main Authors: Hermilo Santiago-Benito, Diana-Margarita Córdova-Esparza, Juan Terven, Noé-Alejandro Castro-Sánchez, Teresa García-Ramirez, Julio-Alejandro Romero-González, José M. Álvarez-Alvarado
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Data
Subjects:
Online Access:https://www.mdpi.com/2306-5729/10/7/94
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849733078588588032
author Hermilo Santiago-Benito
Diana-Margarita Córdova-Esparza
Juan Terven
Noé-Alejandro Castro-Sánchez
Teresa García-Ramirez
Julio-Alejandro Romero-González
José M. Álvarez-Alvarado
author_facet Hermilo Santiago-Benito
Diana-Margarita Córdova-Esparza
Juan Terven
Noé-Alejandro Castro-Sánchez
Teresa García-Ramirez
Julio-Alejandro Romero-González
José M. Álvarez-Alvarado
author_sort Hermilo Santiago-Benito
collection DOAJ
description This article introduces a freely available Spanish–Mixtec parallel corpus designed to foster natural language processing (NLP) development for an indigenous language that remains digitally low-resourced. The dataset, comprising 14,587 sentence pairs, covers Mixtec variants from Guerrero (Tlacoachistlahuaca, Northern Guerrero, and Xochapa) and Oaxaca (Western Coast, Southern Lowland, Santa María Yosoyúa, Central, Lower Cañada, Western Central, San Antonio Huitepec, Upper Western, and Southwestern Central). Texts are classified into four main domains as follows: education, law, health, and religion. To compile these data, we conducted a two-phase collection process as follows: first, an online search of government portals, religious organizations, and Mixtec language blogs; and second, an on-site retrieval of physical texts from the library of the Autonomous University of Querétaro. Scanning and optical character recognition were then performed to digitize physical materials, followed by manual correction to fix character misreadings and remove duplicates or irrelevant segments. We conducted a preliminary evaluation of the collected data to validate its usability in automatic translation systems. From Spanish to Mixtec, a fine-tuned GPT-4o-mini model yielded a BLEU score of 0.22 and a TER score of 122.86, while two fine-tuned open source models mBART-50 and M2M-100 yielded BLEU scores of 4.2 and 2.63 and TER scores of 98.99 and 104.87, respectively. All code demonstrating data usage, along with the final corpus itself, is publicly accessible via GitHub and Figshare. We anticipate that this resource will enable further research into machine translation, speech recognition, and other NLP applications while contributing to the broader goal of preserving and revitalizing the Mixtec language.
format Article
id doaj-art-7de6149f19f34ada807e1b34ca62b4ec
institution DOAJ
issn 2306-5729
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series Data
spelling doaj-art-7de6149f19f34ada807e1b34ca62b4ec2025-08-20T03:08:09ZengMDPI AGData2306-57292025-06-011079410.3390/data10070094Mixtec–Spanish Parallel Text Dataset for Language Technology DevelopmentHermilo Santiago-Benito0Diana-Margarita Córdova-Esparza1Juan Terven2Noé-Alejandro Castro-Sánchez3Teresa García-Ramirez4Julio-Alejandro Romero-González5José M. Álvarez-Alvarado6Facultad de Informática, Universidad Autónoma de Querétaro, Av. de las Ciencias S/N, Campus Juriquilla, Querétaro 76230, MexicoFacultad de Informática, Universidad Autónoma de Querétaro, Av. de las Ciencias S/N, Campus Juriquilla, Querétaro 76230, MexicoCentro de Investigación en Ciencia Aplicada y Tecnología Avanzada—Unidad Querétaro, Instituto Politécnico Nacional, Cerro Blanco No. 141, Col. Colinas del Cimatario, Querétaro 76090, MexicoCentro Nacional de Investigación y Desarrollo Tecnológico, Tecnológico Nacional de México, Interior Internado Palmira S/N, Palmira, Cuernavaca 62493, MexicoCentro de Investigación en Ciencia Aplicada y Tecnología Avanzada—Unidad Querétaro, Instituto Politécnico Nacional, Cerro Blanco No. 141, Col. Colinas del Cimatario, Querétaro 76090, MexicoCentro de Investigación en Ciencia Aplicada y Tecnología Avanzada—Unidad Querétaro, Instituto Politécnico Nacional, Cerro Blanco No. 141, Col. Colinas del Cimatario, Querétaro 76090, MexicoFacultad de Ingeniería, Universidad Autónoma de Querétaro, Querétaro 76010, MexicoThis article introduces a freely available Spanish–Mixtec parallel corpus designed to foster natural language processing (NLP) development for an indigenous language that remains digitally low-resourced. The dataset, comprising 14,587 sentence pairs, covers Mixtec variants from Guerrero (Tlacoachistlahuaca, Northern Guerrero, and Xochapa) and Oaxaca (Western Coast, Southern Lowland, Santa María Yosoyúa, Central, Lower Cañada, Western Central, San Antonio Huitepec, Upper Western, and Southwestern Central). Texts are classified into four main domains as follows: education, law, health, and religion. To compile these data, we conducted a two-phase collection process as follows: first, an online search of government portals, religious organizations, and Mixtec language blogs; and second, an on-site retrieval of physical texts from the library of the Autonomous University of Querétaro. Scanning and optical character recognition were then performed to digitize physical materials, followed by manual correction to fix character misreadings and remove duplicates or irrelevant segments. We conducted a preliminary evaluation of the collected data to validate its usability in automatic translation systems. From Spanish to Mixtec, a fine-tuned GPT-4o-mini model yielded a BLEU score of 0.22 and a TER score of 122.86, while two fine-tuned open source models mBART-50 and M2M-100 yielded BLEU scores of 4.2 and 2.63 and TER scores of 98.99 and 104.87, respectively. All code demonstrating data usage, along with the final corpus itself, is publicly accessible via GitHub and Figshare. We anticipate that this resource will enable further research into machine translation, speech recognition, and other NLP applications while contributing to the broader goal of preserving and revitalizing the Mixtec language.https://www.mdpi.com/2306-5729/10/7/94Mixtec languageparallel corpuslow resource languageOCR
spellingShingle Hermilo Santiago-Benito
Diana-Margarita Córdova-Esparza
Juan Terven
Noé-Alejandro Castro-Sánchez
Teresa García-Ramirez
Julio-Alejandro Romero-González
José M. Álvarez-Alvarado
Mixtec–Spanish Parallel Text Dataset for Language Technology Development
Data
Mixtec language
parallel corpus
low resource language
OCR
title Mixtec–Spanish Parallel Text Dataset for Language Technology Development
title_full Mixtec–Spanish Parallel Text Dataset for Language Technology Development
title_fullStr Mixtec–Spanish Parallel Text Dataset for Language Technology Development
title_full_unstemmed Mixtec–Spanish Parallel Text Dataset for Language Technology Development
title_short Mixtec–Spanish Parallel Text Dataset for Language Technology Development
title_sort mixtec spanish parallel text dataset for language technology development
topic Mixtec language
parallel corpus
low resource language
OCR
url https://www.mdpi.com/2306-5729/10/7/94
work_keys_str_mv AT hermilosantiagobenito mixtecspanishparalleltextdatasetforlanguagetechnologydevelopment
AT dianamargaritacordovaesparza mixtecspanishparalleltextdatasetforlanguagetechnologydevelopment
AT juanterven mixtecspanishparalleltextdatasetforlanguagetechnologydevelopment
AT noealejandrocastrosanchez mixtecspanishparalleltextdatasetforlanguagetechnologydevelopment
AT teresagarciaramirez mixtecspanishparalleltextdatasetforlanguagetechnologydevelopment
AT julioalejandroromerogonzalez mixtecspanishparalleltextdatasetforlanguagetechnologydevelopment
AT josemalvarezalvarado mixtecspanishparalleltextdatasetforlanguagetechnologydevelopment