The flexible piezoresistive sensor has a promising application in fields such as human-computer interaction, artificial intelligence, and smart healthcare due to its advantages of easy signal acquisition, fast response speed, and easy integration. Currently, sensitivity and fabrication methods have become important factors determining whether flexible piezoresistive sensors can be widely applied. Therefore, developing flexible piezoresistive sensors with high sensitivity and simple fabrication methods is of great significance.
Recently, the team led by Li Xiaochun at Taiyuan University of Technology published a paper titled “Bubbles-Induced Porous Structure-Based Flexible Piezoresistive Sensors for Speech Recognition” in the journal ACS Appl. Mater. Interfaces. The researchers utilized the interaction between bubbles generated during the heating of ethanol and PDMS to fabricate flexible piezoresistive sensors with a porous structure. This fabrication method does not require low-temperature and low-pressure experimental conditions and is not limited by templates, making it widely applicable.
Moreover, the fabricated piezoresistive sensors exhibit advantages such as high sensitivity (27.6 kPa-1), short response time (800 μs), and good stability (10,000 cycles). When used for speech recognition, these piezoresistive sensors can not only monitor the muscle movements in the throat of the subject during vocalization but also detect the vibrations of the vocal cords. The recognition accuracy of these piezoresistive sensors for speech signals reaches 94.8%. The research results indicate that these piezoresistive sensors have good application prospects in the field of speech recognition.
Figure 1. (A) The fabrication process of the piezoresistive sensor with a porous structure; (B)-(C) Actual images of porous PDMS and PDMS with graphene attached to the surface; (D) Actual image of the piezoresistive sensor.
Figure 2. (A-C) Schematic and actual images of the piezoresistive sensor under no load, light load, and heavy load; (D) Computer simulation results of pressure distribution when the sensor is subjected to different pressures; (E) Experimental results and computer simulation results of relative resistance changes of piezoresistive sensors made with porous PDMS substrates of different densities as pressure is applied.
Figure 3. (A) Signal response caused by muscle movements when the piezoresistive sensor is connected to the throat; (B-C) Signal responses corresponding to different letters and different pitches of the same letter; (D) Repeated recognition of the Chinese word “nihao”; (E) Recognition of the Chinese characters “ni” and “hao”; (F) Real-time signal changes corresponding to the recitation of Chinese poetry; (G) Visual representation of poetry recitation; (H) Signal response of the piezoresistive sensor while reading poetry at different loudness levels; (I) Signal response of the piezoresistive sensor when 10 subjects recite poetry; (J) Voice signal frequency when 10 subjects read poetry.
Figure 4. (A) Schematic diagram of signal acquisition for vocalization and silent reading; (B) Signal responses caused by silent reading and vocalizing the number “0”; (C) First derivative results of silent and vocal signals; (D) Detailed structure of the CNN model; (E-F) Recognition accuracy and loss function graphs for two datasets; (G) Confusion matrix of speech recognition results.
This paper presents the fabrication of flexible piezoresistive sensors with a porous structure through the interaction of bubbles generated during the heating of ethanol and PDMS. This fabrication method does not require low-temperature and low-pressure experimental conditions and is not limited by templates, making it widely applicable. The fabricated piezoresistive sensors exhibit high sensitivity (27.6 kPa-1), fast response (800 μs), and excellent stability (10,000 cycles). When used for speech recognition, these sensors can monitor both the muscle movements in the throat of the subject during vocalization and the vibrations of the vocal cords. The recognition accuracy of these piezoresistive sensors for speech signals reaches 94.8%. The research results indicate that these piezoresistive sensors have good application prospects in the field of speech recognition.
The first author of this paper is Dr. Gao Xiaoguang, a young teacher at the School of Biomedical Engineering, Taiyuan University of Technology. Master’s student Yuan Lin from the class of 2020 is the second author. Dr. Gao Xiaoguang, Associate Professor Meng Xuejuan, and Professor Li Xiaochun are co-corresponding authors, with Taiyuan University of Technology as the only corresponding institution.
Reference: https://doi.org/10.1021/acsami.3c18233
City University of Hong Kong in “AEM”: Graphene-based moisture-absorbing films inspired by camel hair for passive air cooling with ultra-high cooling power
Shandong University in “Carbon”: PVDF/PMMA/GNP nanocomposites with good compressive mechanical properties, hydrophobicity, and flame retardancy
Jiangnan University in “Nano Letters”: Seconds-level synthesis! Amphiphilic molecules self-spreading for ultra-fast preparation of highly stable zinc-iodine battery cathodes
Source: The article is from the ACS AMI website, organized and edited by Material Analysis and Applications.
Long press to identify the QR code in the image and click to follow
Copyright and Disclaimer:
① All works marked “Material Analysis and Applications” on this website are copyright owned by Material Analysis and Applications. Without authorization, they may not be reproduced, excerpted, or used in other ways. Works authorized by this website should be used within the scope of authorization and must indicate “Source: Material Analysis and Applications”. Violators will be held legally responsible.
② Works marked “Source: xxx (not by this website)” are reprinted from other media. The purpose of reprinting is to convey more information and does not represent this website’s endorsement of its views or its authenticity. This website does not bear direct or joint responsibility for such works’ infringement. If other media, websites, or individuals download and use from this website, they must retain the indicated “source of the manuscript” and assume legal responsibilities such as copyright.
③ If there are issues regarding the content or copyright of the works, please contact this website within three days from the date of publication, otherwise, it will be regarded as a waiver of related rights.
