High-Performance Strain-Sensing Fibers via a Multidimensional Percolation Network Strategy for Intelligent Speech System
Flexible strain sensors often face trade-offs among sensitivity, sensing range, resolution, and stability under deformation. Here, we report a multidimensional percolation strategy for high-performance strain-sensing fibers by integrating 0D Ag microparticles, 1D carbon nanotubes, and 2D carbon sheets into a ternary conductive network through wet spinning. The fibers exhibit a maximum gauge factor of 4640.1, a sensing range up to 65%, and a strain resolution of 0.025%, which is retained under 20% prestrain. Their microstrain sensitivity enables the detection of mechanical signals associated with magnetic microrobot locomotion and the identification of bionic insect model movements using a biomimetic spider-web network. Coupling the fibers with deep learning yields 98.7% recognition accuracy across 21 spoken-word classes. Integration with virtual reality further demonstrates a speech-driven interface in which recognized speech triggers predefined virtual-character responses. These results provide a route toward high-performance fiber sensors for intelligent human–machine interaction.
Authors
- Jun Liang Wu (ORCID: https://orcid.org/0000-0002-9912-5238)
- Shengxin Xiang (ORCID: https://orcid.org/0000-0003-3540-9204)
- Qiongfeng Shi (ORCID: https://orcid.org/0000-0002-5979-1420)
- Hui Zhang (ORCID: https://orcid.org/0000-0003-2989-7547)
- Zhishui Chen
- Xinkai Xie
- Xiao Wei
- Lei Liu
Institutions
- Southeast University (CN)
Publication Details
- Journal
- Nano Letters
- Published
- 2026-11-21
- DOI
- https://doi.org/10.1021/acs.nanolett.6c04249
- Primary Topic
- Advanced Sensor and Energy Harvesting Materials
- Type
- article
- Field-Weighted Citation Impact
- 0.00