A multimodal dataset annotation workbench for benchmarking humor in African languages — Yoruba, Hausa, and Igbo — capturing speech, visual, and linguistic humor cues from comedy content.
🔒 Access is restricted to authorized annotators.
Contact the project team to request access.