CLEAR Global's mission is to help people get information and be heard, whatever language they speak. Evolved from Translators without Borders (TWB), which continues to exist as a community of over 100,000 volunteer linguists worldwide. CLEAR Global taps into this vast community for its language technology projects, combining human linguistic expertise with AI to build speech and translation tools for underserved languages. Develops open-source ASR, TTS, and MT models for crisis-affected and low-resource languages. Flagship initiative TWB Voice is a voice data collection platform integrated with the TWB linguist community, enabling scalable speech data collection. Pilot phase collected 125+ hours of speech in Hausa, Kanuri, and Shuwa Arabic -- producing the first openly available voice datasets for Kanuri and Shuwa Arabic, and the first humanitarian-domain speech data for all three. Also works on Lingala and Congolese Swahili through the LT4CR and SynVoices projects.
CLEAR Global is conducting this UNICEF WCARO NLP landscape mapping consultancy. TWB Voice directly addresses WCA language gaps -- producing the first open speech datasets for Kanuri and Shuwa Arabic, both WCA crisis languages. LT4CR covers Lingala and Congolese Swahili (DRC priority). The TWB community of 100,000+ linguists is an unmatched resource for language data collection. TWB Voice Playbook provides a replicable methodology for UNICEF country offices. Already a UNICEF partner.
Shehu and Uji are COVID-19 chatbots built by CLEAR Global for Nigeria (Hausa, Kanuri) and DR Congo (Congolese Swahili, Lingala) respectively, enabling health information access in local languages during the pandemic.