Saved in:
Bibliographic Details
Main Authors: Montiel-Vazquez, Edwin C., Cruz, Christian Arzate, Gkikas, Stefanos, Kassiotis, Thomas, Giannakakis, Giorgos, Gomez, Randy
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.11417
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917508882104320
author Montiel-Vazquez, Edwin C.
Cruz, Christian Arzate
Gkikas, Stefanos
Kassiotis, Thomas
Giannakakis, Giorgos
Gomez, Randy
author_facet Montiel-Vazquez, Edwin C.
Cruz, Christian Arzate
Gkikas, Stefanos
Kassiotis, Thomas
Giannakakis, Giorgos
Gomez, Randy
contents Co-speech gestures increase engagement and improve speech understanding. Most data-driven robot systems generate rhythmic beat-like motion, yet few integrate semantic emphasis. To address this, we propose a lightweight transformer that derives iconic gesture placement and intensity from text and emotion alone, requiring no audio input at inference time. The model outperforms GPT-4o in both semantic gesture placement classification and intensity regression on the BEAT2 dataset, while remaining computationally compact and suitable for real-time deployment on embodied agents.
format Preprint
id arxiv_https___arxiv_org_abs_2604_11417
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Efficient Emotion-Aware Iconic Gesture Prediction for Robot Co-Speech
Montiel-Vazquez, Edwin C.
Cruz, Christian Arzate
Gkikas, Stefanos
Kassiotis, Thomas
Giannakakis, Giorgos
Gomez, Randy
Robotics
Artificial Intelligence
Co-speech gestures increase engagement and improve speech understanding. Most data-driven robot systems generate rhythmic beat-like motion, yet few integrate semantic emphasis. To address this, we propose a lightweight transformer that derives iconic gesture placement and intensity from text and emotion alone, requiring no audio input at inference time. The model outperforms GPT-4o in both semantic gesture placement classification and intensity regression on the BEAT2 dataset, while remaining computationally compact and suitable for real-time deployment on embodied agents.
title Efficient Emotion-Aware Iconic Gesture Prediction for Robot Co-Speech
topic Robotics
Artificial Intelligence
url https://arxiv.org/abs/2604.11417