A domain-filtered subset of Kikuyu (Gĩkũyũ) speech data focused on banking and financial-transaction content, combined into a single repository with train, test, and validation splits.

Source

This dataset is a filtered subset of Anv-ke/kikuyu, part of the African Next Voices (ANV) collection. All audio, transcriptions, and underlying speaker data originate from that source dataset. Full credit for data collection belongs to the… See the full description on the dataset page: https://huggingface.co/datasets/NjeriKahoro/anv-kikuyu-banking-subset.