Documenting oral culture on Wikimedia

South Asia is home to more than 800 languages-over 10% of the world’s approximately 7,000 languages. However, much of this cultural knowledge remains absent from the internet. Folk songs, oral histories, proverbs, traditional ecological knowledge, and oral history are often passed down verbally and are at risk of disappearing as languages become endangered and younger generations move away from these practices resulting in shrinking language diversity.

As a speaker of the vulnerable South Asian language Angika, I found Wikimedia projects incredibly useful to digitize the rich cultural heritage of my language and region. From uploading oral history and folk songs on Commons, editing on Angika Wikipedia incubator to creating Wikisource pages that can be used to create citable sources on small-language Wikipedias, I have had first-hand experience of a speaker of an underrepresented language using the internet to visibilize the language’s various knowledge forms. With the realization that South Asia is incredibly linguistically and culturally diverse, I began to look into ways of supporting the same for other languages as well. 

From a pilot to a growing initiative

The oral culture projects have reached their current state based on a project that began in 2021– the Oral Culture Transcription Toolkit – a toolkit for South Asian language speakers who are interested in documenting the oral culture of their language. It was based on two pre-existing toolkits- the Wikitongues toolkit and OpenSpeaks toolkit. It has been translated into multiple languages, edited to include more information based on feedback, and audio-visuals for training topics. 

In 2022, a needs assessment research was conducted to understand what people’s perceived notions about documenting languages and cultures are, the challenges on ground for the same, and support they need for successful documentation and digitization. Based on the input in our interviews and surveys of this research, we planned to train speakers of South Asian languages to conduct fieldwork, to provide them with minimal equipment like tripod for shooting videos, and train them on integrating them on Wikimedia projects. 

Eddie Avila’s (co-managing director-Global Voices) insight on language activists as bridges for emerging language enthusiasts was instrumental in taking the next step of inviting volunteers to receive support for documenting oral culture:”Even if there are not existing activists in one’s language, opportunities for cross-linguistic, cross-regional mentorship between activists from another language can guide and inspire interested individuals.”

These two projects had Nitesh Gill as the communication and community coordinator. Nitesh not only took command of coordinating with project participants, she has also been an advisor and helped plan meetups for subsequent Indic oral culture projects- IOC 2023 and IOC 2024-25.

In the IOC 2023 project we aimed to recruit citizen archivists/participants/volunteers from five language communities. The response exceeded expectations, with participants joining from nine language communities across India, Nepal, and Bangladesh, resulting in documentation efforts in eleven languages. We got participation from Wikimedians, language activists, and linguists from Jawaharlal Nehru University (my alma mater). Mohammad Iflaq was the community coordinator in this project and was also in charge of creating the public-facing Meta pages for the project and running its editathon.

The subsequent IOC 2024-25 project brought together participants from across India, Nepal, and Pakistan. We introduced in-person meetings, collaborated with an academic expert to strengthen documentation quality, and began researching tools to support documentation and digitization efforts. We succeeded in engaging 23 participants in this year-long project, representing 15 South Asian languages. This project’s coordinator, Yakshitha Poojary innovated in designing engaging activities for participants during the in-person meetups. 

Wikitongues founder Daniel Borge Udell has remained an advisor since the toolkit, in addition to that the Wikitongues organization has served as a fiscal sponsor for the 2025 project. Praveen Das has provided valuable insights into building partnerships since 2024 and also helped establish partnership with the Taaoos Chaman project. Maarti Maitreyi (Whose Knowledge?) is serving as an advisor and an observer in the 2026 iteration. 

Impact

The participants of the Indic oral culture projects have recorded around 381 oral culture videos in 22 languages so far. Many of these participating language communities had little prior engagement with Wikimedia projects. Oral culture documentation became an entry point for broader representation of languages in Wikimedia open knowledge initiatives.

Several of these languages are underrepresented online and the languages and their oral culture are under the risk of disappearance:

Participants have documented a wide range of cultural practices. In Bangladesh, contributors recorded marriage folk songs that are now remembered only by older generations. The Tulu community documented O Bele songs- songs associated with farming traditions that are increasingly disappearing as younger generations move away from agriculture. Across communities, contributors have documented folk songs, stories, oral histories, medicinal knowledge, riddles, proverbs, a game, and ritual practices that previously had little or no digital presence.

The ongoing ‘Documenting and disseminating Indic oral culture on Wikimedia projects’ (IOC 2026) initiative includes participants from India, Nepal, Zimbabwe, and Nigeria, reflecting the broader relevance of oral culture documentation among linguistically diverse communities around the world.

• Endangered/vulnerable languages out of the languages involved in the projects so far:
  • Level of endangerment as per UNESCO list and Ethnologue list: Torwali (definitely endangered), Tulu (vulnerable), Mising (definitely endangered), Kharia (vulnerable), Deori (definitely endangered), Sora (vulnerable), Kui (vulnerable), Bodo (vulnerable), Mundari (vulnerable), Kodava (endangered), Godwari (endangered), Kurukh (endangered)

Building a community of practice and learnings

One of the most important learnings from the project has been that documenting oral culture is not simply a matter of training individuals. It requires building a community of practice. Many questions had to be explored through experimentation. Who would be interested in participating? What kinds of training would be useful? How frequently should sessions be held? What documentation methods would work across different cultural contexts? How could contributors be supported beyond the initial training period? Over time, we found that native speakers, language activists, linguistics students, Wikimedians, and citizen archivists brought complementary strengths to the work. Through workshops, discussions, field documentation, and peer learning, participants began sharing not only techniques but also challenges and solutions.

Perhaps the most significant lesson has been that there is no universal model for documenting oral culture. Each language community has its own cultural context, priorities, and challenges. At the same time, certain principles have emerged as broadly applicable. Successful documentation efforts require local leadership, sustained mentoring, flexible training structures, and opportunities for contributors to learn from one another. They also require recognizing that oral culture documentation is both technical and relational work. Recording knowledge depends not only on equipment and platforms, but also on trust, community relationships, and cultural sensitivity.

Map demonstrating spread and variety of the Indic oral culture documented (sample videos only)- https://meta.wikimedia.org/wiki/Indic_Oral_Culture_Projects/Map 

Oral culture documentation in the age of AI

The rise of artificial intelligence has renewed conversations about the sources of knowledge available online. Much of the world’s linguistic and cultural diversity remains underrepresented in digital spaces, limiting what future technologies can learn from and represent.

Projects focused on oral culture documentation contribute something that cannot be generated automatically: original, community-based knowledge gathered through direct engagement with people and cultural practices. Recording a folk song from its last practitioners, documenting local ecological knowledge, or preserving oral histories requires human relationships, fieldwork, and contextual understanding.

Looking ahead

As the oral culture projects grow, we have initiated the Wikimedia Oral culture working group. It is a global working group that brings together ideas, discussions, and challenges around oral culture with regards to Wikimedia projects. Wikimedia projects have been used widely for representation of various knowledge forms, they have served as a platform for digitizing oral culture as well. The purpose of this group is to bring diverse knowledge forms and cultural practices that are recited and passed down generations orally- oral history, folk songs, folk tales, indigenous medicinal knowledge etc. on to Wikimedia projects to enrich them. To discuss the current challenges in digitizing oral culture, potential solutions, current and evolving practices, and ideas around using Wikimedia projects innovatively for the same.

This working group is open for Wikimedia globally, if you wish to become a part of it, sign up here- Wikimedia Oral Culture Working Group