A Prompt-Guided and Quality-Aware Robust Text–Audio Intent Recognition Framework for Elderly Care
Sensors · 18.08.2026 · 26 (16) · 5233
DOI: 10.3390/s26165233 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, online erschienen am 18.08.2026. Verfasst von Zhimin Wei und 5 weiteren Autor:innen. Beteiligte Einrichtung: College of Mechanical Engineering, Beihua University, Jilin 132021, China. Hauptthemen laut MEDLINE: Intention, Pattern Recognition, Automated. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Jilin Province Science and Technology Department, Department of Education of Jilin Province.
Abstract (Originalsprache)
In natural language understanding, intent recognition plays a central role in human–computer interaction. However, in elderly-care scenarios, acoustic signals are often affected by atypical speech patterns, slower speaking rates, and environmental noise, making audio information less reliable and reducing the effectiveness of conventional text–audio fusion methods. To address this problem, we propose a prompt-guided and quality-aware text–audio intent recognition framework. Specifically, a χ2-based intent prototype soft prompt is introduced to enhance the semantic representation of text. Then, a residual-free text-guided cross-attention module is designed to refine degraded acoustic features using textual semantics as reliable guidance. In addition, a dynamic fusion gate is developed to adjust the contributions of text and audio based on modality reliability and intent-related information. Experiments on the MIntRec dataset with simulated acoustic degradation show that the proposed model achieves 60.90% accuracy, 60.80% weighted F1, and 57.90% macro-F1, outperforming several competitive baselines. These results indicate that the proposed framework can improve the robustness of intent recognition under challenging acoustic conditions in elderly-oriented interaction scenarios.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- College of Mechanical Engineering, Beihua University, Jilin 132021, China.
ORCID-Profile: Zhimin Wei, Yao Wang, Xiaolong Zhou
Förderung
- Jilin Province Science and Technology Department
- Department of Education of Jilin Province
Zitieren
APA
Wei, Z., Tian, S., Wang, Y., Wang, Y., Zhou, X., & Li, J. (2026). A Prompt-Guided and Quality-Aware Robust Text–Audio Intent Recognition Framework for Elderly Care. Sensors, 26(16), 5233. https://doi.org/10.3390/s26165233
Vancouver
Wei Z, Tian S, Wang Y, Wang Y, Zhou X, Li J. A Prompt-Guided and Quality-Aware Robust Text–Audio Intent Recognition Framework for Elderly Care. Sensors. 2026;26(16):5233. doi:10.3390/s26165233
BibTeX
@article{Wei2026,
author = {Wei, Zhimin and Tian, Shuhao and Wang, Yanzhen and Wang, Yao and Zhou, Xiaolong and Li, Jianyong},
title = {A Prompt-Guided and Quality-Aware Robust Text–Audio Intent Recognition Framework for Elderly Care},
journal = {Sensors},
year = {2026},
volume = {26},
number = {16},
pages = {5233},
doi = {10.3390/s26165233}
}
EUVIMED