Restricted — review the source terms before use.
Every column the registry tracks. Values marked derivedwere filled by a rule or read from the source's text rather than stated in its metadata; the rule or the quoted sentence is shown under the value.
Record indexed from Hugging Face
27 Oct 2026
Source status: Active
27 Oct 2026
This is the same multilingual corpus, sliced for other languages the platform indexes it under.
Multi-language MT benchmark/dataset covering 46 African languages + Arabic + French + English. Afar (aar) is one of the 46 supported languages. Total rows: t…
Data format: json
Data format: json
Data format: json
Data format: json
Data format: json
Data format: json
Data format: json
Data format: json
Data format: json
Data format: json
Data format: json
Data format: json
African language MT dataset from UBC including Igbo
Dataset card explicitly maps kau to Kanuri; language-specific example count not stated in the retrieved source.
Data format: json
Hugging Face dataset tagged for Kinyarwanda. This may be a broad multilingual resource rather than a language-specific corpus.
Data format: json
Data format: json
Data format: json
Extensive parallel textual translation corpus compiling over 31,000 verified sentences mapping the standardized literary orthography rules of the Fon (Fɔngbè…
Spoken audio database tracking language code fon (Language Number 2801), archiving continuous native spoken scripts, \"Good News\" audio lessons, and dramati…
Descriptive grammatical index and structural morphosyntactic logs outlining dialect boundaries, phonetic variations, and genealogical language classification…