ORAL
ORAL - kriOl(u) laRge lAnguage modeLs
Description

The project is the result of a strategic partnership between Iscte – Lisbon University Institute and the University of Cape Verde (Uni-CV), under the Multilingualism pillar, and its main objective is to promote the digital transformation of the Cape Verdean language (kea) through the development of computational linguistic resources and advanced artificial intelligence technologies. The initiative addresses the lack of natural language processing tools for Cape Verdean Creole, making all results available under an open-access and open-source model, thereby contributing to the digital inclusion of the more than 1.5 million speakers of the language in Cape Verde and the diaspora.

The project aims to develop a pioneering set of technological resources, including text and speech corpora, large-scale language models (LLMs), bidirectional machine translation systems between Kea and European Portuguese, automatic speech recognition, chatbots, voicebots, phonetic glossaries, and application programming interfaces (APIs) that will enable third parties to develop new applications and services based on the Cape Verdean language. These resources will be essential for integrating Kea into the international language technology ecosystem, filling a gap that currently exists in major global AI platforms.

In addition to the technological component, the project includes significant investment in the training and capacity-building of teachers, researchers, students, civil servants, businesspeople, and other members of society, promoting the adoption of the tools developed and strengthening digital skills related to multilingualism. At the same time, it will contribute to the promotion, standardization, and future official recognition of the Cape Verdean language, supporting Cape Verde’s public policies and strengthening its presence on the Internet and in digital services.

 

The collaboration between Iscte and Uni-CV is based on an institutional relationship that began in 2018 and brings together complementary expertise in linguistics, artificial intelligence, machine learning, and speech technologies. The project benefits from the international expertise of both institutions, involving leading researchers in the fields of computational linguistics and generative AI, as well as advanced high-performance computing infrastructure provided by Iscte. This combination of scientific knowledge and technological capability positions the consortium as a leader in applied research on Creole languages and strengthens academic cooperation between Portugal and Cape Verde.

The project’s results are expected to have a significant impact on education, public administration, health, tourism, e-commerce, and digital services, promoting greater accessibility, inclusion, and innovation. By developing language technologies of high scientific value for a historically underrepresented language, the project will help reduce digital inequalities, stimulate technological entrepreneurship, and establish the Cape Verdean language as an integral part of the global digital transformation.

Internal Partners
Research Centre Research Group Role in Project Begin Date End Date
ISTAR-Iscte Digital Living Spaces Partner 2025-10-06 2026-12-31
External Partners
Institution Country Role in Project Begin Date End Date
Cape Verde University (UniCv) Cape Verde Partner 2025-10-06 2026-12-31
Project Team
Name Affiliation Role in Project Begin Date End Date
Miguel Sales Dias Professor Catedrático (DTDA); Integrated Researcher (ISTAR-Iscte); Researcher 2025-10-06 2026-12-31
Project Fundings

No records found.

Publication Outputs

No records found.

Related Research Data Records

No records found.

Related References in the Media

No records found.

Other Outputs

No records found.

Project Files

No records found.

ORAL - kriOl(u) laRge lAnguage modeLs
2025-10-06
2026-12-31