cs.CLMar 15, 2026

Vavanagi: a Community-run Platform for Documentation of the Hula Language in Papua New Guinea

Authors: Bri Olewale, Raphael Merx, Ekaterina Vylomova

Organizations: Vula’a Kunenai Community, Central Province, Papua New Guinea · The University of Melbourne, Melbourne, Australia

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

We present Vavanagi, a community-run platform for Hula (Vula'a), an Austronesian language of Papua New Guinea with approximately 10,000 speakers. Vavanagi supports crowdsourced English-Hula text translation and voice recording, with elder-led review and community-governed data infrastructure. To date, 77 translators and 4 reviewers have produced over 12k parallel sentence pairs covering 9k unique Hula words. We also propose a multi-level framework for measuring community involvement, from consultation to fully community-initiated and governed projects. We position Vavanagi at Level 5: initiative, design, implementation, and data governance all sit within the Hula community, making it, to our knowledge, the first community-led language technology initiative for a language of this size. Vavanagi shows how language technology can bridge village-based and urban members, connect generations, and support cultural heritage on the community's own terms.

Figures & tables

Figure 1: Overview of the Vavanagi platform and Hula language
Figure 2: Community involvement spectrum, from consultation to community governance. See mapping of prior work onto the spectrum in Table 1 .
WorkIDOGLevel
Petrariu et al. [ 22 ] – AromanianEEEE2
Li et al. [ 23 ] – KhinalugEEEE2
Prabowo et al. [ 24 ] – Meher/WoirataEECS3
Cox et al. [ 19 ] – Tsuut’inaESSS4
Frontull et al. [ 25 ] – LadinSSCS4
Krajinovic et al. [ 10 ] – NafsanSSCS4
Table 1: Mapping of prior work onto the community involvement spectrum. Columns: I nitiative origin, D esign authority, O perational leadership, data G overnance. Codes: E = externally led, S = shared/co-shaped, C = community-led.
Figure 3: Workflow on the Vavanagi platform, with one admin user, 77 translators, and 4 reviewers.
Figure 4: Screenshots of the Vavanagi platform showing the home page, translator data-entry view, and reviewer interface.
MetricValue
Sentence pairs12,124+
Unique English words7,948
Unique Hula words9,556
Median sentence length8 words / 39 characters
Approved after 1 / 2 / 3+ translations91% / 8% / 1%
Translators77
Table 2: Vavanagi corpus and participation statistics.
More in this topic

Low-Resource Language Processing

Low-resource language processing builds and evaluates language technologies for languages with limited data, tools, or digital representation.

See all 565 papers

Explore similar work

CardsList

  1. Towards Model as a Library: Offline, Community-Sourced AI for Low-Resource African Languages

    Sep 29, 2026Fendji K. E. Jean LouisSpeech ProcessingLow-Resource Language Processing

    Centre of Research, Experimentation and Production SCEMI, University of Ngaoundere, Ngaoundere, Cameroon · Stellenbosch Institute for Advanced Study Wallenberg Research Centre at Stellenbosch University, Stellenbosch, South Africa

  2. A Survey of Text and Speech Resources for Hausa and Fongbe: Availability, Quality, and Gaps for NLP Development

    Apr 13, 2026Mahounan Pericles Adjovi, Victor Olufemi, Roald Eiselen +1Low-Resource Language ProcessingSpeech Processing

    Carnegie Mellon University Africa, Kigali, Rwanda · Centre for Text Technology, North-West University, South Africa

  3. Designing Explainable Conversational Agentic Systems for Guaraní Speakers

    Mar 5, 2026Samantha Adorno, Akshata Kishore Moharir, Ratna KandalaSpoken Dialogue SystemsHuman-Centered AI

    University of Kansas, Lawrence, Kansas, USA · Independent Researcher, Sherwood, Oregon, USA