SGGNet$^2$: Speech-Scene Graph Grounding Network for Speech-guided Navigation

TL;DR

This paper introduces SGGNet$^2$, a robust speech-scene graph grounding network that improves spoken language understanding for robot navigation by leveraging acoustic similarities and integrating ASR systems.

Contribution

The paper presents a novel extension of the scene-graph grounding network that incorporates acoustic similarity from ASR to enhance speech grounding accuracy.

Findings

Effective grounding of spoken commands demonstrated.

Improved navigation performance on real robot.

Robustness to speech variability shown.

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$^2$) 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$^2$ in a speech-based navigation task using a real quadruped robot, RBQ-3, from Rainbow Robotics.