alffa_wolof
15,118 Wolof ASR samples
Official site consultedAI: datasets
Last verified: 1 October 2026
Wolof speech recognition dataset: audio recordings with transcriptions, 15,118 rows (14,000 train, 1,120 dev) for 2.07 GB. Files are in Parquet format.
Services
- Audio, speaker ID, transcription, filename
Access requirements
Publicly available on Hugging Face; license not specified.
Sources
- alffa_wolof (new tab), verified on 1 October 2026Official site consulted