Deep‐Learning‐Enabled MXene‐Based Artificial Throat: Toward Sound Detection and Speech Recognition

Jin, Yukun; Wen, Bo; Gu, Zixiong; Jiang, Xiantao; Shu, Xiaolan; Zeng, Zhenping; Zhang, Yupeng; Guo, Zhinan; Chen, Yun; Zheng, Tingting; Yue, Yutao; Zhang, Han; Ding, Huijun

doi:10.1002/admt.202000262

Cited by 51 publications

(38 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Lastly,t here are still many other opportunities for MXene-basedm aterials for use in different applications beyondt he exciting potentiala se nergy storagem aterials. For example, EMI shielding, [134] light heat, [135] 3D printing, [136] machine learning, and artificial intelligence [137][138][139] etc.,w hich may be influenced by the intercalation process.…”

Section: Discussionmentioning

confidence: 99%

Interlayer Space Engineering of MXenes for Electrochemical Energy Storage Applications

Tang

Huang

Qiu

et al. 2020

Chemistry A European J

View full text Add to dashboard Cite

The increasing demand for high‐performance rechargeable energy storage systems has stimulated the exploration of advanced electrode materials. MXenes are a class of two‐dimensional (2D) inorganic transition metal carbides/nitrides, which are promising candidates in electrodes. The layered structure facilitates ion insertion/extraction, which offers promising electrochemical characteristics for electrochemical energy storage. However, the low capacity accompanied by sluggish electrochemical kinetics of electrodes as well as interlayer restacking and collapse significantly impede their practical applications. Recently, interlayer space engineering of MXenes by different chemical strategies have been widely investigated in designing functional materials for various applications. In this review, an overview of the most recent progress of 2D MXenes engineering by intercalation, surface modification as well as heterostructures design is provided. Moreover, some critical challenges in future research on MXene‐based electrodes have been also proposed.

show abstract

Section: Discussionmentioning

confidence: 99%

Interlayer Space Engineering of MXenes for Electrochemical Energy Storage Applications

Tang

Huang

Qiu

et al. 2020

Chemistry A European J

View full text Add to dashboard Cite

show abstract

“…For the vowel "o", the signal shows a slow increase before the peak, a slow decrease after the peak, and a sharp peak, which may be attributed to a high tone but a small mouth opening, as shown in Figure 7e. For the vowel "u", the high tone results in a sharp peak and fast amplitude decrease, and the long diphthong pronunciation results in a mild amplitude upswing, [54] as shown in Figure 7f. The HM is also applied for voice recognition of English words, including "laser" (pronounced as/"leɪzer/) and "sensor" (pronounced as/"sensər/), as shown in Figure 8.…”

Section: Applications Of the Hm-based Flexible Sensormentioning

confidence: 97%

“…When the tester speaks a certain vowel, obvious signal differences can be found due to the difference in how muscles contract around the vocal cords. For the vowel, "a" (pronounced as/əɪ/), the small mouth opening and gentle vocal cord vibration result in a low peak amplitude, [54] as shown in Figure 7b. For the vowel "e" (pronounced as/i:/), two sharp peaks can be found due to the long pronunciation and high tone, [46] which indicates an obvious pressure variation, as shown in Figure 7c.…”

Section: Applications Of the Hm-based Flexible Sensormentioning

confidence: 99%

Laser Direct Writing of Highly Ordered Two‐Level Hierarchical Microstructures for Flexible Piezoresistive Sensor with Enhanced Sensitivity

Zhang

Chen

Xiao

et al. 2021

Adv Materials Inter

View full text Add to dashboard Cite

With the development of micro‐nano manufacturing technology, various hierarchical microstructures (HMs) are used to improve the sensitivity and measurement range of flexible pressure sensors. However, the fabrication of highly ordered HMs in simple, fast, and low‐cost ways remains a great challenge. In this work, laser direct writing technology is used to fabricate highly ordered HMs to enhance the sensitivity of flexible piezoresistive sensors. The HMs show a lateral expansion with the increasing pressure due to good flexibility and small force bearing areas, resulting in more significant change in contact area than the single‐level microstructures, which leads to an enhanced sensitivity. Two case studies are conducted to verify the performance of the sensor with laser processed HMs. Experimental results show that the pulsating blood pressure signal of radial artery can be detected by attaching the sensor on the wrist. When the sensor is attached on the neck, it can also detect the vibration signal of vocal cord when speaking. These results successfully demonstrate the potential of laser processing in fabricating highly ordered HMs to achieve highly sensitive flexible piezoresistive sensors for various applications, such as wearable health monitoring and human–computer interactions.

show abstract

“…The fuzzy logic and ANFIS were applied for flood prediction [9]. However, compared to these "shallow" modelling methods, the deep learning (DL) models have been developed as a reliable estimation and prediction tool in many fields, such as image classification [10], speech recognition [11], COVID-19 diagnostic and prognostic analysis [12], rainfall-runoff modeling [12], and streamflow prediction [13]. Among the multiple DL models, long short-term memory (LSTM) and convolutional neural networks (CNN) are the two most commonly used models.…”

Section: Introductionmentioning

confidence: 99%

Comparison of Forecasting Models for Real-Time Monitoring of Water Quality Parameters Based on Hybrid Deep Learning Neural Networks

Sha

Zhang

et al. 2021

Water

View full text Add to dashboard Cite

Accurate real-time water quality prediction is of great significance for local environmental managers to deal with upcoming events and emergencies to develop best management practices. In this study, the performances in real-time water quality forecasting based on different deep learning (DL) models with different input data pre-processing methods were compared. There were three popular DL models concerned, including the convolutional neural network (CNN), long short-term memory neural network (LSTM), and hybrid CNN–LSTM. Two types of input data were applied, including the original one-dimensional time series and the two-dimensional grey image based on the complete ensemble empirical mode decomposition algorithm with adaptive noise (CEEMDAN) decomposition. Each type of input data was used in each DL model to forecast the real-time monitoring water quality parameters of dissolved oxygen (DO) and total nitrogen (TN). The results showed that (1) the performances of CNN–LSTM were superior to the standalone model CNN and LSTM; (2) the models used CEEMDAN-based input data performed much better than the models used the original input data, while the improvements for non-periodic parameter TN were much greater than that for periodic parameter DO; and (3) the model accuracies gradually decreased with the increase of prediction steps, while the original input data decayed faster than the CEEMDAN-based input data and the non-periodic parameter TN decayed faster than the periodic parameter DO. Overall, the input data preprocessed by the CEEMDAN method could effectively improve the forecasting performances of deep learning models, and this improvement was especially significant for non-periodic parameters of TN.

show abstract

Deep‐Learning‐Enabled MXene‐Based Artificial Throat: Toward Sound Detection and Speech Recognition

Cited by 51 publications

References 35 publications

Interlayer Space Engineering of MXenes for Electrochemical Energy Storage Applications

Interlayer Space Engineering of MXenes for Electrochemical Energy Storage Applications

Laser Direct Writing of Highly Ordered Two‐Level Hierarchical Microstructures for Flexible Piezoresistive Sensor with Enhanced Sensitivity

Comparison of Forecasting Models for Real-Time Monitoring of Water Quality Parameters Based on Hybrid Deep Learning Neural Networks

Contact Info

Product

Resources

About