Elsevier

Nano Energy

Volume 113, August 2023, 108534
Nano Energy

Anti-interference self-powered acoustic fabric for complex acoustic environments

https://doi.org/10.1016/j.nanoen.2023.108534Get rights and content

Highlights

  • Low requirement for sound source: regardless of whether wearing a mask, or the vocal volume in large or small.

  • Resistance to propagation process interference: performs well under the conditions of windstorm and rainstorm.

  • High quality sound recording: sensitive, accurate and reliable with wide range, high SNR, small drift and good stability.

Abstract

Traditional airborne microphones are at risk of failure due to their dependence on airborne media in complex acoustic environments (CAEs). Here, we report an anti-interference self-powered acoustic fabric (ASAF) that can shield the interfering factors in the process of sound production and propagation to serve as a precise and wearable sound receiver. The use of the soft and safe woven structure polyvinylidene fluoride (PVDF) as a vibration-sensitive layer enables the ASAF to record human speech at wide vibration frequencies (0–5000 Hz). A speech recognition system is established which can recognize 25 words related to extreme weather conditions, with more than 95.8% accuracy. This speech recognition is carried out in CAEs such as wearing masks, silent communication, windstorms, and rainstorms, with corresponding losses of accuracy less than 1.6%, 6.7%, 6.3%, and 5.8%, respectively. The ASAF is expected to facilitate outdoor rescuers, journalists, students, and other professionals working in CAEs.

Introduction

Sound is an important tool for information transmission [1]. Human voices can be efficiently transmitted to a certain place or many listeners through sound waves in ambient air, without the need for direct touch. However, traditional air-conducting microphones face the challenge of disabling under complex acoustic environment (CAE) [2]. The CAE could cause severe damage to the sound quality at the sound source and arouse distortions in the sound propagation process. The former may limit the loudness and clarity of a person’s voice while the latter will introduce significant background noise and environmental absorption losses [3], [4]. Meanwhile, the situations faced by different people in life are ever-changing and the above CAE scenarios are commonly encountered in life [5]. For example, students wearing masks in offices, libraries, and museums may be unable to resist the urge to whisper, and the private whispers between a couple of lovers are never wanted to be heard. Besides, masks on the face also make people sound slurred when they are on the phone [6]. Interference during sound propagation is even more ubiquitous. Everyone may be in stadiums, square gatherings, and workshops, which are noisy and cannot filter out the interference by simple and easy means because the sound sources are uncountable [7]. Windstorms, rainstorms, and sandstorms due to the violent air movement, not only can produce great background noise but also directly impact the microphone to produce huge interference [8]. However, an outdoor journalist has to hold a bulky microphone to try to make the voice captured in such huge interference. There is an urgent need to find a wearable microphone that can withstand harsh CAE conditions. Thus, wearable acoustic sensors have several technical requirements:[9] (i) high sensitivity, wide response frequency, and low detection limit, with a high-precision level of voice perception; (ii) Non-air medium conduction, but directly attached to the skin to collect vibration signals, to avoid the interference in the process of external propagation; (iii) High mechanical flexibility and robust, so that the equipment can adhere to the human body conformally and operate stably under long-term dynamic movement.

Recently, resistive vibration sensors based on microcracked films have shown increased sensitivity to subtle strains [10], [11], [12], [13], [14]. However, due to the weak adhesion of the crack layer, it is easy to peel off from the flexible substrate during cyclic high-frequency measurements, and the continuous growth of cracks causes performance drift, which poses serious problems for its long-term working stability. Capacitive vibration sensors have a flat frequency response and low power consumption [6], [15], but like resistive sensors, they continuously consume energy and may be affected by heat accumulation. Triboelectric nanogenerators (TENGs) do not require an external power supply, and their simple and reliable structures can be well-designed as skin-friendly acoustic sensors [16], [17], [18]. Wang’s paper-based triboelectric nanogenerator [19], and self-powered triboelectric auditory [20], are classical works in recording sound and harvesting wave energy. Chen’s group reported a waterproof acoustic sensor (WAS) as a wearable translation interface to communicate with machines [21]. Lee’s group discusses a new platform of speaker recognition with Gaussian Mixture Model (GMM) which is a typical work for artificial intelligence [22]. The performance of triboelectric devices is especially conducive to quickly data collecting, which has unique advantages for establishing speech recognition systems [23].

Here, we report a triboelectric-based anti-interference self-powered acoustic fabric (ASAF) that resists the effects of CAE. At the sound source, it can record ambiguous speech and enable silent communication, while during the sound propagation, it will not be interfered by extreme weather. Therefore, it can serve as a precise, wearable acoustic receiver for office workers, students, outdoor journalists, and other occupations. The ASAF uses a simple polyvinylidene fluoride (PVDF) braided fabric as the vibration-sensitive layer, greatly reducing the difficulty of equipment manufacturing and maintenance [24]. With the aid of the material’s flexibility and braided structure, the ASAF not only can be worn comfortably on the skin surface near the thyroid cartilage for a long time but also can detect sounds up to 5000 Hz. Around the basic frequency of the human voice, it has a signal-to-noise ratio (SNR) of more than 61.1 dB, and 35.1 dB above 1778 Hz, with less than 0.0045% signal drift. The ASAF also demonstrates extremely high stability. After 6 × 105 cycles, it is placed for 30 days, and the output performance remains nearly the same (only lowing 4.4%) for another 4 × 105 cycles. Excellent performance enables ASAF to meet the requirements (i)–(iii) of wearable acoustic sensors at the same time, achieving reliable work in (a) silent communication, (b) wearing masks, resistance to airflow noise such as (c) windstorms and (d) rainstorms in extreme weather. We establish a 25-word speech recognition system for the needs of our users, such as outdoor journalists and rescue workers, with a recognition accuracy of more than 95.8%. Even in the above CAE conditions, the accuracy is only reduced by 1.6%, 6.7%, 6.3%, and 5.8%, respectively. In comparison, the signal differences recorded by the reference microphone between the ideal environment and CAE are more than 50%, which fully illustrates the incompetence of the microphone under CAE conditions.

Section snippets

Results

The ASAF attaches slightly below the thyroid cartilage and effectively converts vibrations into electrical signals. The PVDF is cut into strips and woven into a vibration-sensitive layer and sandwiched between the copper electrodes on both sides to form a two-electrode structure TENG, as shown in Fig. 1A. The optical photographs are shown in Fig. S1A–B. Compared to flat structures, the woven structure has better flexibility and sensitivity [25], [26]. The PVDF has good mechanical ability with a

Conclusion

In the way of collecting sound vibration information from the throat, the ASAF’s voice acquisition ability is greatly enhanced and provided a solution to the problem that traditional air conduction microphones are incompetent in CAE conditions. Taking advantages of the self-powered TENG, we have developed a fully biologically safe ASAF. The soft woven PVDF film as a vibration-sensitive layer can enable the ASAF to cover a wide frequency range (0–5000 Hz) and record precisely (0.0045% drift and

Manufacture of the ASAF

The PVDF was cut into thin strips of 1 mm width, woven in a plain pattern with every 10 strips in the warp and weft directions. The copper foil was cut into 15 × 15 mm2 and sandwich the PVDF in the middle. The device was then encapsulated by PE.

Characterization and measurement

The mechanical deformation process was controlled by Micro-tester 5948 (Instron). The current, voltage and charge were measured by a 6514-system electrometer (Keithley). The signals higher than 5000 Hz were measured on 2 channel digital storage

CRediT authorship contribution statement

Jizhong Zhao: Conceptualization, Methodology, Formal analysis, Investigation, Writing – original draft. Yao Yuan: Methodology, Software, Formal analysis. Wentao Lei: Formal analysis, Investigation, Writing – original draft. Li Zhao: Conceptualization. Andeng Liu: Investigation. Meidan Ye: Supervision. Jianyang Wu: Supervision. Shihui Guo: Supervision. Wenxi Guo: Writing – review & editing, Supervision.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work was supported by the Science and Technology Project of Fujian Province (2022H0001), the National Natural Science Foundation of China (22175146), the Guangdong Natural Science Foundation (2021A1515010680), the Fundamental Research Funds for the Central Universities of China (20720210027), the “111 Project” (B16029).

Jizhong Zhao is a Ph.D. candidate in College of Physical Science and Technology, Xiamen University, China. His research focuses on the fabrication of triboelectric nanogenerators serving as wearable sensors, energy harvesting and storage from environment, and preparation and application of hydrogels.

References (50)

  • X. Yan et al.

    Experimental studies on the rain noise of lightweight roofs: natural rains vs artificial rains

    Appl. Acoust.

    (2016)
  • H. Ding et al.

    Recent advances in nanomaterial-enabled acoustic devices for audible sound generation and detection

    Nanoscale

    (2019)
  • S. Lee et al.

    An ultrathin conformable vibration-responsive electronic skin for quantitative vocal recognition

    Nat. Commun.

    (2019)
  • M.S. Longuet-Higgins

    An analytic model of sound production by raindrops

    J. Fluid Mech.

    (2006)
  • X.F. Huang et al.

    Application of university campus noise map based on noise propagation model: a case in Guangxi University

    Sustainability

    (2022)
  • S. Lee et al.

    An electret-powered skin-attachable auditory sensor that functions in harsh acoustic environments

    Adv. Mater.

    (2022)
  • Q.S. Yang et al.

    Mixed-modality speech recognition and interaction using a wearable artificial throat

    Nat. Mach. Intell.

    (2023)
  • G. Dutilleux et al.

    Comparing sound emergence and sound pressure level as predictors of short-term annoyance from wind turbine noise

    Acta Acust.

    (2020)
  • T.S. Dinh et al.

    Ultrasensitive anti-interference voice recognition by bio-inspired skin-attachable self-cleaning acoustic sensors

    ACS Nano

    (2019)
  • B. Park et al.

    Dramatically enhanced mechanosensitivity and signal-to-noise ratio of nanoscale crack-based sensors: effect of crack depth

    Adv. Mater.

    (2016)
  • D. Kang et al.

    Ultrasensitive mechanical crack-based sensor inspired by the spider sensory system

    Nature

    (2014)
  • L. Zhao et al.

    Highly stretchable, adhesive, and self-healing silk fibroin-dopted hydrogels for wearable sensors

    Adv. Health Mater.

    (2021)
  • S. Lee et al.

    Skin-attachable acoustic sensor for realizing auditory electronic skin

    Adv. Mater.

    (2022)
  • Z. Yu et al.

    Nanoporous PVDF hollow fiber employed piezo-tribo nanogenerator for effective acoustic harvesting

    ACS Appl. Mater. Interfaces

    (2021)
  • J. Yang et al.

    Eardrum-inspired active sensors for self-powered cardiovascular system characterization and throat-attached anti-interference voice recognition

    Adv. Mater.

    (2015)
  • Cited by (2)

    Jizhong Zhao is a Ph.D. candidate in College of Physical Science and Technology, Xiamen University, China. His research focuses on the fabrication of triboelectric nanogenerators serving as wearable sensors, energy harvesting and storage from environment, and preparation and application of hydrogels.

    Yuan Yao received master’s degree in Materials Science and Engineering from Qingdao university of science and technology in 2022. Then he joined Prof. Shihui Guo’s group at the School of Information, Xiamen University for his Ph.D. Currently, his research interests focus on the flexible sensors for mixed reality, senseless intelligent wearable devices and computer graphics related to human dynamic reconstruction.

    Wentao Lei is currently a master’s student in the College of Physical Science and Technology, Xiamen University. His major is photoelectric information engineering. The main research interest is TENG for acoustic sensing and protection against corrosion.

    Li Zhao received her master’s degree in condensed matter physics in College of Physical Science and Technology, Xiamen University, China, 2021. Her research focuses on the preparation, analysis and application of hydrogels, especially composite hydrogels with a variety of properties based on biomaterials.

    Andeng Liu is a Master candidate in School of Chemistry and Chemical Engineering at Xiamen University. His research interests are focused on the mesoscopic regulation of silk and its sensing properties.

    Meidan Ye received her Ph.D. in College of Chemistry and Chemical Engineering at Xiamen University in 2014. She then joined Research Institute for Soft Matter and Biomimetics, Department of Physics, College of Physical Science and Technology at Xiamen University as an associate Professor in 2014. She was promoted to Professor in 2021. Her research interests are on multi-functional materials for flexible devices, i.e., electrochemical energy storage devices and wearable sensors.

    Dr. Jianyang Wu is currently a Professor in Department of Physics from Xiamen University (XMU), China. He studied his Ph.D. in Nanomechanics from Norwegian University of Science and Technology, Norway. His research interest mainly focused on the relationships of microstructures-property-technology of smart soft matter and low-dimensional materials. Recently, his particular interests mainly concentrated on the key scientific concerns of unconventional energy resources.

    Shihui Guo is currently an Associate Professor of Xiamen University. He obtained his bachelor’s degree from Yuanpei College of Peking University in 2010 and Ph.D. from National Computing Animation Center, Bournemouth University in 2015. In 2019, he received the fellowship of Leaders in Innovation from Chinese Academy of Engineering and Royal Academy of Engineering (UK). He has worked as a postdoctoral researcher in the research group of Professor Nadia Thalmann, a member of Swiss Academy of Engineering. His research interests mainly focus on human-computer interaction and computer graphics. More than 40 papers have been published in international top conferences and journals, such as ACM CHI, TVCG, ISMAR, CVPR and TIP.

    Wenxi Guo received his B.S. and Ph.D. degree in College of Chemistry and Chemical Engineering at Xiamen University in 2008 and 2014, respectively. Then, he had been an assistant research fellow in the group of Professor Caofeng Pan at Beijing Institute of Nanoenergy and Nanosystems, Chinese Academy of Sciences. He then joined College of Physical Science and Technology, Xiamen University and became an associate professor since 2015. His research focuses on the fields of flexible thin film solar cells, physicochemical properties of silk fibroin and flexible intelligent sensing system.

    1

    Jizhong Zhao and Yuan Yao contributed equally to this work.

    View full text