SGGNet$^2$: Speech-Scene Graph Grounding Network for Speech-guided Navigation
Abstract
The spoken language serves as an accessible and efficient interface, enabling non-experts and disabled users to interact with complex assistant robots. However, accurately grounding language utterances gives a significant challenge due to the acoustic variability in speakers' voices and environmental noise. In this work, we propose a novel speech-scene graph grounding network (SGGNet) that robustly grounds spoken utterances by leveraging the acoustic similarity between correctly recognized and misrecognized words obtained from automatic speech recognition (ASR) systems. To incorporate the acoustic similarity, we extend our previous grounding model, the scene-graph-based grounding network (SGGNet), with the ASR model from NVIDIA NeMo. We accomplish this by feeding the latent vector of speech pronunciations into the BERT-based grounding network within SGGNet. We evaluate the effectiveness of using latent vectors of speech commands in grounding through qualitative and quantitative studies. We also demonstrate the capability of SGGNet in a speech-based navigation task using a real quadruped robot, RBQ-3, from Rainbow Robotics.
Keywords
Cite
@article{arxiv.2307.07468,
title = {SGGNet$^2$: Speech-Scene Graph Grounding Network for Speech-guided Navigation},
author = {Dohyun Kim and Yeseung Kim and Jaehwi Jang and Minjae Song and Woojin Choi and Daehyung Park},
journal= {arXiv preprint arXiv:2307.07468},
year = {2024}
}
Comments
7 pages, 6 figures, Published at 2023 IEEE International Conference on Robot and Human Interactive Communication (RO-MAN), [Dohyun Kim, Yeseung Kim, Jaehwi Jang, and Minjae Song] contributed equally to this work