Research, Language, and Cultural Heritage
Tonalli Corpus brings together linguistic research, artificial intelligence, and community engagement to support the documentation and study of Nahuatl and Totonac languages.
Beyond the development of a multilingual parallel corpus, the project has produced scientific publications and educational and audiovisual materials that help communicate its findings to researchers, educators, students, and the wider public.
This section presents the project’s academic contributions, the Ancestral Code comic book, and documentary videos produced during fieldwork in Indigenous communities in Puebla, Mexico.
Scientific Publications
1. Spectral Robustness in Low-Resource Language Identification: A Comparative Deep Learning Study of Nahuatl and Mexican Spanish
Authors: G. Lomeli, R. Alvarez-Sanchez, D. Pinto, F. Perez-Tellez, E. Varela, and S. Paniagua Rivera
Year: 2027
Publisher: Springer
Publication series: Communications in Computer and Information Science (CCIS)
Volume: 3097
Publication status: Published
DOI: https://doi.org/10.1007/978-3-032-37441-7_24
About the publication
This study examines deep learning approaches to distinguishing Nahuatl from Mexican Spanish using spectral characteristics of speech. It contributes to research on automatic language identification in low-resource linguistic settings and explores approaches that may support future language technologies.
Bibliographic reference
Lomeli, G., Alvarez-Sanchez, R., Pinto, D., Perez-Tellez, F., Varela, E., & Paniagua Rivera, S. (2027). Spectral robustness in low-resource language identification: A comparative deep learning study of Nahuatl and Mexican Spanish. Communications in Computer and Information Science, 3097. Springer. https://doi.org/10.1007/978-3-032-37441-7_24
Series information
-
Publisher: Springer Nature
-
Series: Communications in Computer and Information Science
-
Print ISSN: 1865-0929
-
Electronic ISSN: 1865-0937
2. Automatic Language Identification of Indigenous Mexican Languages: A Transfer Learning Approach for Nahuatl and Totonaco
Authors: R. Alvarez-Sanchez, G. Lomeli, D. Pinto, F. Perez-Tellez, E. Varela, and S. Paniagua Rivera
Year: 2027
Publisher: Springer
Publication series: Communications in Computer and Information Science (CCIS)
Volume: 3097
Publication status: Published
DOI: https://doi.org/10.1007/978-3-032-37441-7_27
About the publication
This research explores the application of transfer learning to automatic language identification for Nahuatl and Totonac. The study investigates how existing deep learning approaches can be adapted to languages with limited computational resources, contributing to the development of more inclusive speech technologies.
Bibliographic reference
Alvarez-Sanchez, R., Lomeli, G., Pinto, D., Perez-Tellez, F., Varela, E., & Paniagua Rivera, S. (2027). Automatic language identification of Indigenous Mexican languages: A transfer learning approach for Nahuatl and Totonaco. Communications in Computer and Information Science, 3097. Springer. https://doi.org/10.1007/978-3-032-37441-7_27
Series information
-
Publisher: Springer Nature
-
Series: Communications in Computer and Information Science
-
Print ISSN: 1865-0929
-
Electronic ISSN: 1865-0937
3. TonalliCorpus: Towards the Construction of a Multilingual Parallel Corpus of Nahuatl and Totonac Variations for Further AI Applications
Journal: Computación y Sistemas
Acceptance date: May 16, 2026
Scheduled issue: Volume 30, Number 4 (2026)
Publication status: Accepted for publication
About the publication
This article presents the development of Tonalli Corpus, a multilingual linguistic resource incorporating Nahuatl and Totonac speech, transcriptions, and Spanish and English translations.
It describes the methodology used to collect and organize the linguistic materials and discusses their potential applications in natural language processing, language documentation, and artificial intelligence.
The publication contributes to efforts to improve the availability of computational resources for Indigenous Mexican languages.
The final bibliographic reference and DOI will be added when the article is published. See here the acceptation letter (CyS_constancia_David_Pinto)
Ancestral Code — The Comic Book
An Adventure into the Treasure of Indigenous Languages
Ancestral Code (Código ancestral) is an illustrated adventure created by Dr. David Pinto, principal investigator of Tonalli Corpus, as a personal initiative to bring the project’s ideas to a wider audience.
Through a fictional journey in search of a mysterious treasure, the story introduces readers to the cultural and linguistic heritage of Indigenous communities. The adventure ultimately leads to the Tonalli Corpus website, connecting the fictional story with a real research initiative dedicated to documenting Nahuatl and Totonac languages.
The comic book was conceived as an accessible way to engage students, teachers, families, and younger audiences. One of its main objectives is to share the story with the communities where the Tonalli Corpus recordings were collected, encouraging interest in their languages and cultural heritage.
Watch the Official Teaser and Trailer
Discover the world of Ancestral Code through two promotional videos introducing the comic book and its adventure.
Official Teaser
A first glimpse into the world of Ancestral Code.
Official Trailer
Explore the adventure and discover the story behind Ancestral Code.
Read the Comic Book
Follow the adventure and discover where the search for the ancestral treasure leads.
Read the Digital Comic Book:
https://heyzine.com/flip-book/eb8c53dec5.html
Publication Details
-
Title: Ancestral Code (Código ancestral)
-
Author and Creator: Dr. David Pinto
-
Format: Illustrated comic book
-
Edition: Preliminary digital edition
-
ISBN: Registration in progress
-
Related Project: Tonalli Corpus
The comic book is an independent educational and cultural outreach initiative developed by the project’s principal investigator. Plans include making it available to the participating communities as a complementary resource for promoting interest in Indigenous languages.
Documentary Videos
Voices, Communities, and Fieldwork
The Tonalli Corpus project involved visits to Nahuatl- and Totonac-speaking communities across Puebla, Mexico.
In addition to collecting linguistic recordings, the research team documented aspects of its fieldwork through photography and video.
These audiovisual materials provide a complementary perspective on the project, highlighting the communities visited and the activities involved in building the corpus.
The first three documentary videos are presented below.
Teteles de Ávila Castillo
Location: Teteles de Ávila Castillo, Puebla, Mexico
A documentary associated with the Tonalli Corpus fieldwork activities in Teteles de Ávila Castillo.
Watch the documentary:
https://youtu.be/FTmpTj8J4a0
Nanacatlán
Location: Nanacatlán, Puebla, Mexico
A documentary associated with the project’s fieldwork and linguistic documentation activities in Nanacatlán.
Watch the documentary:
https://youtu.be/FzlbGnv10z0
San José Miahuatlán
Location: San José Miahuatlán, Puebla, Mexico
A documentary associated with the Tonalli Corpus recording activities in San José Miahuatlán.
Watch the documentary:
https://youtu.be/vuqvwFKbwQQ
Additional Documentaries
Additional documentary materials may be published as further audiovisual recordings from the project are prepared.
These videos form part of the project’s broader effort to communicate its work beyond academic publications and make its activities more accessible to participating communities and the public.