With the rapid development of human–machine interfaces, artificial acoustic sensors play an important role in the hearing impaired. Here, an ultrathin eardrum‐like triboelectric acoustic sensor (ETAS) is presented consisting of silver‐coated nanofibers, whose thickness is only 40 µm. The sensitivity and frequency response range of the ETAS are closely related to the geometric parameters. The ETAS endows a high sensitivity of 228.5 mV Pa−1 at 95 dB, and the ETAS has a broad frequency response ranging from 20 to 5000 Hz, which can be tuned by adjusting the thickness, size, or shape of the sensor. Cooperating with artificial intelligence (AI) algorithms, the ETAS can achieve real‐time voice conversion with a high identification accuracy of 92.64%. Under good working property and the AI system, the ETAS simplifies signal processing and reduces the power consumption. This work presents a strategy for self‐power auditory systems, which can greatly accelerate the miniaturization of self‐powered systems used in wearable electronics, augmented reality, virtual reality, and control hubs for automation.

Air pollution is a major global problem, closely related to economic and social development and ecological environment construction. Air pollution data for most regions of China have a close correlation with time and seasons and are affected by multidimensional factors such as meteorology and air quality. In contrast with classical peaks-over-threshold modeling approaches, we use a deep learning technique and three new dynamic conditional generalized Pareto distribution (DCP) models with weather and air quality factors for fitting the time-dependence of the air pollutant concentration and make statistical inferences about their application in air quality analysis.Specifically, in the proposed three DCP models, a dynamic autoregressive exponential function mechanism is applied for the time-varying scale parameter and tail index of the conditional generalized Pareto distribution, and a sufficiently high threshold is chosen using two threshold selection procedures. The probabilistic properties of the DCP model and the statistical properties of the maximum likelihood estimation (MLE) are investigated, simulating and showing the stability and sensitivity of the MLE estimations. The three proposed models are applied to fit the PM2.5 time series in Beijing from 2015 to 2021. Real data are used to illustrate the advantages of the DCP, especially compared to the estimation volatility of GARCH and AIC or BIC criteria. The DCP model involving both the mixed weather and air quality factors performs better than the other two models with weather factors or air quality factors alone. Finally, a prediction modelbased on long short-term memory (LSTM) is used to predict PM2.5 concentration, achieving ideal results.