eJSL - solo

This dataset contains 78 distinct utterances, each performed by one male and one female signer across the six Ekman's basic emotions (anger, disgust, fear, joy, sadness, surprise) and the neutral state, yielding 1,092 MP4 clips in total. The signers are native JSL (Japanese Sign Language) speakers who work as vocational deaf actors and can read and write fluently in Japanese. All instructions and utterances were textually presented to the signers in Japanese. The 78 utterances were adopted from a public transcript with substantial modifications in consultation with a professional sign language interpreter. The primary intended use of this dataset is benchmarking for emotion recognition in sign language speakers.