DOLMA-NLP Joins Meta's Language Technology Partner Program to Advance Middle Eastern Language Technologies 🎉
We are thrilled to announce that DOLMA-NLP has been selected to join Meta’s Language Technology Partner Program, an initiative supporting UNESCO’s work to promote linguistic diversity and inclusivity in AI models, particularly for underserved languages.
Our Contribution
DOLMA-NLP brings to this partnership our recently developed multilingual parallel corpus across 8 Middle Eastern languages. This comprehensive dataset, enhanced with audio recordings, positions us to make immediate contributions to speech-to-speech translation applications.
Partnership Objectives
Through this collaboration with Meta, DOLMA-NLP aims to:
- Leverage Meta’s expertise and resources to build effective speech recognition and machine translation models for underserved Middle Eastern languages
- Create standardized digital resources as foundations for future language technology development
- Establish sustainable collaboration frameworks between language communities, researchers, and technology developers
- Collect additional data for more languages in the region
- Increase visibility and enable communities with resources provided by Meta for deployment
Alignment with Our Mission
This partnership perfectly aligns with DOLMA-NLP’s core mission of providing advanced AI resources for underserved language communities. It supports our commitment to linguistic diversity that transcends regional divisions while accelerating language digitalization efforts. By participating in this program, we will help bridge the digital divide while preserving the unique heritage of Middle Eastern languages. Consistent with our open-source priority initiative, all language technologies developed through this program will be made freely available to foster collaboration, transparency, and community involvement.
We look forward to contributing to Meta’s efforts in supporting underserved languages and bringing more speakers into the digital conversation.