Auditory systems are the most efficient and direct strategy for communication between human beings and robots.In this domain,flexible acoustic sensors with magnetic,electric,mechanical,and optic foundations have attra...Auditory systems are the most efficient and direct strategy for communication between human beings and robots.In this domain,flexible acoustic sensors with magnetic,electric,mechanical,and optic foundations have attracted significant attention as key parts of future voice user interfaces(VUIs)for intuitive human–machine interaction.This study investigated a novel machine learning-based voice recognition platform using an MXene/MoS_(2) flexible vibration sensor(FVS)with high sensitivity for acoustic recognition.The performance of the MXene/MoS_(2) FVS was systematically investigated both theoretically and experimentally,and the MXene/MoS_(2) FVS exhibited high sensitivity(25.8 mV/dB).An MXene/MoS_(2) FVS with a broadband response of 40–3,000 Hz was developed by designing a periodically ordered architecture featuring systematic optimization.This study also investigated a machine learning-based speaker recognition process,for which a machine-learning-based artificial neural network was designed and trained.The developed neural network achieved high speaker recognition accuracy(99.1%).展开更多
基金supported by the National Natural Science Foundation of China(Nos.51972025,61888102,and 62174152)the Young Elite Scientists Sponsorship Program by the China Association for Science and Technology(CAST)(No.2018QNRC001)+1 种基金the Strategic Priority Program of the Chinese Academy of Sciences(No.XDA16021100)the Science and Technology Development Plan of Jilin Province(No.20210101168JC).
文摘Auditory systems are the most efficient and direct strategy for communication between human beings and robots.In this domain,flexible acoustic sensors with magnetic,electric,mechanical,and optic foundations have attracted significant attention as key parts of future voice user interfaces(VUIs)for intuitive human–machine interaction.This study investigated a novel machine learning-based voice recognition platform using an MXene/MoS_(2) flexible vibration sensor(FVS)with high sensitivity for acoustic recognition.The performance of the MXene/MoS_(2) FVS was systematically investigated both theoretically and experimentally,and the MXene/MoS_(2) FVS exhibited high sensitivity(25.8 mV/dB).An MXene/MoS_(2) FVS with a broadband response of 40–3,000 Hz was developed by designing a periodically ordered architecture featuring systematic optimization.This study also investigated a machine learning-based speaker recognition process,for which a machine-learning-based artificial neural network was designed and trained.The developed neural network achieved high speaker recognition accuracy(99.1%).
