ABSTRACT
In this article, we present three studies involving WhatsHap, a mobile system designed to deliver speech as vibrations on the forearm with minimal hardware demands and practice time. After only 4.2 h of training on a 24-haptic phoneme vocabulary and on how to combine these to form words, participants were able to generalize their phoneme identification skills to the understanding of untrained English words, correctly identifying 65% of words in phrases rendered with a user-controlled interval between words, and up to 59% with a fixed interval. Ultimately, participants were able to complete 88% of simple communicative tasks that elicited spontaneous speech and semi-structured bidirectional conversation using the apparatus. We conclude by providing insights as to how such a system may ultimately be used for communication under more natural conditions.
