English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

This paper presents a novel framework for speech-driven gesture production,applicable to virtual agents to enhance human-computer interaction.Specifically, we extend recent deep-learning-based, data-driven methods forspeech-driven gesture generation by incorporating representation learning. Ourmodel takes speech as input and produces gestures as output, in the form of asequence of 3D coordinates. We provide an analysis of different representationsfor the input (speech) and the output (motion) of the network by both objectiveand subjective evaluations. We also analyse the importance of smoothing of theproduced motion. Our results indicated that the proposed method improved on ourbaseline in terms of objective measures. For example, it better captured themotion dynamics and better matched the motion-speed distribution. Moreover, weperformed user studies on two different datasets. The studies confirmed thatour proposed method is perceived as more natural than the baseline, althoughthe difference in the studies was eliminated by appropriate post-processing:hip-centering and smoothing. We conclude that it is important to take bothmotion representation and post-processing into account when designing anautomatic gesture-production method.

Moving fast and slow: Analysis of representations and post-processing in speech-driven automatic gesture generation