Generalizing a model for animating adverbs of manner in American Sign Language

This work aims to show that a model produced to generate adverbs of manner can be generalized and applied to a variety of neutral animated signs for avatar sign language synthesis. This paper presents the generalization of a new approach that was first presented at SLTAT 2019 in Hamburg for modeling language processes that manifest themselves as modifications to the visual-manual channel. This work discusses extensions for generalizability to the model to be effective for a broader range of signs including one-handed and two-handed signs, repeating and non-repeating signs, signs with contact, and additional rotational adjustments to the wrists. This paper also includes interim results from an ongoing user study.