Privacy Policy

Birtal Lab is a community-led African AI research initiative based in Kampala, Uganda. We create speech and language data for Sudanese and other underrepresented African languages. For privacy questions, contact privacy@birtallab.org.

This policy covers the public website (birtallab.org) and the contributor platform (accounts, onboarding, and the annotation workspace). It explains what we collect, why, how we protect it, and the rights you hold over your data.

Application information you give when asking to take part: your name, email address, phone number if you give one, the country and city you live in, an age range, how far you studied, and what you do. We ask for an age range, never a date of birth, and we do not ask where you were born.

Account information you provide: email address, display name, preferred language, the languages and dialects you can contribute, and avatar choice.

Contributions: audio recordings you submit, transcriptions, annotations, preference judgments, and the rationale you write.

Onboarding & quality data: review outcomes.

Technical & integrity data: authentication tokens, device/browser metadata, and anti-cheat telemetry used solely to protect data quality.

Communications: messages you send us.

We do not deliberately collect special-category data (e.g., health, religion). Please do not include it in recordings unless it is part of the language sample and you consent to its open publication.

To create your account and operate the platform (contract); to build and validate language datasets and conduct research (your consent and our legitimate research interest); to maintain security and data integrity (legitimate interest); and to comply with law.

To run the contributor workflow, compute quality and tiers, produce research and datasets, and improve our methods. We do not sell your personal data, and we do not use it for advertising.

Before publication, contributions are processed in line with your consent and the Data Rights & Consent terms. Published datasets are designed to represent the *language*, not to expose contributors’ identities; direct identifiers are not part of published corpora.

We share data only with service providers that help us operate, under contract: Supabase (database, file storage, authentication), Google Cloud (hosting), and Speechmatics and Featherless (which process audio for transcription and response generation). We may disclose data if required by law. We never sell personal data.

Data is stored with our infrastructure providers using access controls and row-level security. Audio is held in secured storage. We keep account data while your account is active and contribution data as needed for research integrity; you may request deletion (Section 10), subject to the limits explained in Data Rights & Consent for data already published openly.

Our providers may process data outside your country (for example, provider regions in the EU or US). We rely on appropriate contractual and technical safeguards for such transfers.

Subject to applicable law, you may access, correct, export, or delete your personal data, object to or restrict certain processing, and withdraw consent at any time. Withdrawal does not affect processing already carried out (explained plainly in Data Rights & Consent). To exercise rights, contact privacy@birtallab.org.

The platform is for contributors aged 18 and over. The application offers no age range below 18, and we do not knowingly collect data from anyone below this age.

We will post updates here and adjust the "Last updated" date; material changes will be highlighted.

Birtal Lab, Kampala, Uganda — privacy@birtallab.org.