New Interfaces and Approaches to Machine Learning When Classifying Gestures within Music

Interactive music uses wearable sensors (i.e., gestural interfaces—GIs) and biometric datasets to reinvent traditional human–computer interaction and enhance music composition. In recent years, machine learning (ML) has been important for the artform. This is because ML helps process complex biometric datasets from GIs when predicting musical actions (termed performance gestures). ML allows musicians to create novel interactions with digital media. Wekinator is a popular ML software amongst artists, allowing users to train models through demonstration. It is built on the Waikato Environment for Knowledge Analysis (WEKA) framework, which is used to build supervised predictive models. Previous research has used biometric data from GIs to train specific ML models. However, previous research does not inform optimum ML model choice, within music, or compare model performance. Wekinator offers several ML models. Thus, we used Wekinator and the Myo armband GI and study three performance gestures for piano practice to solve this problem. Using these, we trained all models in Wekinator and investigated their accuracy, how gesture representation affects model accuracy and if optimisation can arise. Results show that neural networks are the strongest continuous classifiers, mapping behaviour differs amongst continuous models, optimisation can occur and gesture representation disparately affects model mapping behaviour; impacting music practice.     

Mechanistic model for the dielectric spectrum of a simple dielectric material

ABSTRACT

In this paper, we perform molecular dynamics simulations of a dielectric fluidic material composed of permanent molecular dipoles. The dielectric spectrum features two peaks at lower frequencies than the system phonon frequency. The primary peak is observed at all temperatures studied and shifts toward lower frequencies as the temperature decreases. During this shift, the secondary peak emerges with a higher peak frequency than the primary peak. The secondary peak amplitude increases as the temperature decreases. Both peaks are dependent on the wavevector; in the small wavevector regime, the primary peak is shifted to higher frequencies as the wavevector squared and the secondary peak amplitude increases as the wavevector increases, but shows no shift in frequency. From the polarisation balance equation, we propose a model for the dielectric spectrum. This captures the spectrum features, and we conjecture that the primary peak is due to dipole moment correlations (Debye-type) and the secondary peak is due to the correlation between the dipole moment and a microscopic local field.     

动态压缩下Zr基非晶合金失效释能机理

为研究Zr 基非晶合金动态压缩条件下的失效释能机理,采用力学试验机、霍普金森杆、高速摄影、差示扫描量热分析(differential scanning calorimetry, DSC)、扫描电镜(scanning electron microscope, SEM)等,得到了材料应力应变曲线、高速摄影图像、失效式样微观形貌及DSC 曲线,根据实验数据计算了材料的晶化激活能,并拟合了材料的JH-2(Johnson-Holmquist Ⅱ)模型,对材料动态失效过程进行有限元数值模拟。实验结果表明,压缩条件下材料为脆性断裂,断口处观察到典型的脉状纹样及液滴状结构,材料失效过程伴随着释能现象;数值模拟结果表明,材料裂纹局部的瞬时内能大于材料晶化激活能。动态压缩下材料的失效释能机理即为材料破碎释放储存的弹性势能,并导致材料局部晶化释能,释能强度与应变率成正相关。     

Irreversible capacity of the amorphous silicon thin-film electrodes

Irreversible processes accompanying the lithium incorporation into amorphous thin-film silicon are investigated. It is shown that the irreversible processes occurred during the cathodic polarization result in the formation of passive film at the silicon surface. The passive film at silicon is close, in its composition, to the passive film at carbonaceous materials. However, unlike the carbonaceous electrodes, no effect of electrolyte composition on the irreversible capacity of the silicon electrodes is observed.     

超声波辅助Ru-B非晶态合金常压催化肉桂醛加氢制备肉桂醇

采用KBH₄化学还原法制备了超细Ru-B非晶态合金催化剂。Ru-B催化剂在肉桂醇液相常压加氢中显示出较高的肉桂醇选择性。在加氢过程中使用超声辐射能大大提高反应速率而肉桂醇的选择性几乎保持不变。加氢速率随着超声频率或超声时间的增加而增加。通过XRD、XPS、TEM、BET和ICP等各种表征手段，简要研究了超声辐射对Ru-B催化剂结构和电子特征的影响。同时，还讨论了超声对肉桂醛选择性加氢制肉桂醇催化性能的促进作用。     

由非晶态Ni-P低温制备Ni2P催化剂及其对二苯并噻吩加氢脱硫的催化性能

 以非晶态Ni-P合金为前驱体,在低温下通过PH3处理制备了Ni2P/SiO2-Al2O3催化剂,并用X射线衍射(XRD)、 透射电镜(TEM)、 电感耦合等离子体发射光谱、 N2吸附和X射线光电子能谱(XPS)进行了表征,以二苯并噻吩为探针,在小型连续流动固定床反应器上考察了催化剂的加氢脱硫性能. XRD结果表明,在200～300 ℃范围内前驱体都可以完全转化为Ni2P, 随着磷化温度的升高,晶体结构变得越来越完整. TEM观察发现, Ni-P粒子和Ni2P粒子的平均尺寸都在40～50 nm, 并且都能够高分散在SiO2-Al2O3载体上. XPS结果表明,不论是非负载还是SiO2-Al2O3负载的Ni2P, 表层主要为Ni2P和钝化层Ni3(PO4)2. Ni2P/SiO2-Al2O3催化剂在实验范围内表现出很好的二苯并噻吩加氢脱硫性能.     

Infrared ellipsometry of interdiffusion in thin films of miscible polymers

A new application of infrared ellipsometry is reported. Specifically, the interdiffusion between thin films of miscible polymers—poly(methyl methacrylate) and poly(vinylidene fluoride)—is detected in a non‐invasive measurement. A novel technique of data analysis for interdiffusion was developed and is described. The validity of the approach is supported by simulations of diffusion in a bilayer. The onset of extensive interdiffusion over a time period of 15 min occurs at a temperature of 160 °C. At a temperature of 190 °C, the data show that complete mixing of a bilayer (850 nm thick) occurs within 30 s, which is consistent with previously reported values of the mutual diffusion coefficient. Infrared ellipsometry is non‐invasive, applicable at elevated temperatures and relatively fast and sensitive. Although, in these measurements, it was unable to determine a concentration profile at the interface, infrared ellipsometry was used successfully to detect when interdiffusion had occurred. Hence, it is a useful means for screening polymer pairs for miscibility. Copyright © 2004 John Wiley & Sons, Ltd.     

电化学合成CoP-Cu复合丝及其巨磁阻抗效应

电化学合成CoP-Cu复合丝及其巨磁阻抗效应;钴磷非晶合金;电沉积;巨磁阻抗;复合丝     

Liquid-Phase Hydrogenation of Chloronitrobenzene Over Pt-Sn-B Amorphous Catalyst Supported by Carbon Nanotubes

The Pt-Sn-B/carbon nanotubes (CNTs) catalyst was prepared by impregnation-chemical reduction method. Its catalytic performance was evaluated by liquid-phase hydrogenation of chloronitrobenzene (CNB). The results showed that the catalyst had higher catalytic performance than common hydrogenation catalysts. The conversion of CNB could reach 99.9%, and the dechlorination of chloroaniline (CAN) was less than 1.9% when catalyzed by Pt-Sn-B/CNTs and more than 8.0% when catalyzed by common hydrogenation catalysts. X-ray diffraction and selected area electron diffraction analysis showed that Pt-Sn-B/CNTs had an amorphous alloy structure that can improve catalytic performance. Transmission electron micrograph image showed that the catalyst particles were highly distributed on the surface of CNTs. The hydrodechlorination of CNB was mainly affected by the unique structure of CNTs and the nature of the amorphous metals on the surface of CNTs. The relationship between the interaction of CNTs and amorphous metals and the catalytic performance of the catalyst is also discussed. Translated from the Chinese Journal of Catalysis, 2005, 26(3) (in Chinese)     

非晶态Fe－W合金镀层的表面改性

将Ｆｅ－Ｗ非晶态镀层在铬酸盐钝化液中进行化学钝化与电化学钝化，使之形成含铬钝化膜。阳极极化曲线说明，钝化后镀层的孔蚀电位向正向移动了约１．６８Ｖ；Ｆｅ－Ｗ非晶镀层在氯化钠溶液中浸泡近１ｈ表面发生严重腐蚀，而经钝化处理的镀层浸泡１个月，表面无变化，仍具有金属光泽。