As a major component of speech signal processing, speech emotion recognition has become increasingly essential to understanding human communication. Benefitting from deep learning, many researchers have proposed vario...As a major component of speech signal processing, speech emotion recognition has become increasingly essential to understanding human communication. Benefitting from deep learning, many researchers have proposed various unsupervised models to extract effective emotional features and supervised models to train emotion recognition systems. In this paper, we utilize semi-supervised ladder networks for speech emotion recognition. The model is trained by minimizing the supervised loss and auxiliary unsupervised cost function. The addition of the unsupervised auxiliary task provides powerful discriminative representations of the input features, and is also regarded as the regularization of the emotional supervised task. We also compare the ladder network with other classical autoencoder structures. The experiments were conducted on the interactive emotional dyadic motion capture (IEMOCAP) database, and the results reveal that the proposed methods achieve superior performance with a small number of labelled data and achieves better performance than other methods.展开更多
The theory of RC uniform ladder networks based upon the recurrence of voltage and cur-rent functions is extended as a vehicle to analyse the dynamic characteristics of reg lines. Meth-ods for computing the time consta...The theory of RC uniform ladder networks based upon the recurrence of voltage and cur-rent functions is extended as a vehicle to analyse the dynamic characteristics of reg lines. Meth-ods for computing the time constants and simplifying the transfer functions for reg lines are alsopresented.展开更多
In order to improve the multiplexing capability of the optical sensors based on the lower interferential optic fiber sensing technology and the white light fiber-optic Mach-Zehnder interferometer,reflective ladder top...In order to improve the multiplexing capability of the optical sensors based on the lower interferential optic fiber sensing technology and the white light fiber-optic Mach-Zehnder interferometer,reflective ladder topology network ( RLT) with tailored formula was proposed. The topology network consists of 6 rungs sensing elements linked by 5 couplers. Two cases with different choices of couplers were contrasted: one is equal coupling ratio,and the other is tailored coupling ratio. Through the simulation of these two cases,the detailed multiplexing capability was analyzed,and accordingly the experiments were also carried out. The simulation results showed that,the tailored formula enhances the multiplexing capability of the structure. In the first case, the maximum number of sensors which can be multiplexed is 8,and in the other case is 12 fiber optic sensors. The experimental results have a good agreement with numerical simulation results. Thus,it is considered expedient to incorporate RLT into large-scale building,grounds,bridges,dams,tunnels,highways and perimeter security.展开更多
The title complex, {[Cu2（4,4＇-bipyridine）2（μ-O2CMe）2（O2CMe）2],H2O}n 1, was synthesized and structurally characterized by X-ray crystallography. It crystallizes in monoclinic, space group C2/c with a = 13.4474...The title complex, {[Cu2（4,4＇-bipyridine）2（μ-O2CMe）2（O2CMe）2],H2O}n 1, was synthesized and structurally characterized by X-ray crystallography. It crystallizes in monoclinic, space group C2/c with a = 13.4474（5）, b = 11.7566（2）, c = 19.5380（6）A, β = 92.930（2）°, V = 3084.84（16） A^3, Z = 4, Cu2C28N409H30, Mr = 693.64, Dc = 1.494 g/cm^3, F（000） = 1424 and μ（MoKα） = 1.436 mm^-1. With the use of 2062 observed reflections （I 〉 2σ（I））, the structure was refined to R = 0.0769 and wR = 0.2154. In complex 1, the dimeric copper acetate units are linked through 4,4’-bipyridine to yield ID molecular ladders. These ladders are connected via O-H…O hydrogen bonds to generate 2D layers, which are further linked through C-H…O hydrogen bonds to give a 3D supramolecular network.展开更多
基金supported by National Natural Science Foundation of China(Nos.61425017 and 61773379)the National Key Research&Development Plan of China(No.2017YFB1002804)
文摘As a major component of speech signal processing, speech emotion recognition has become increasingly essential to understanding human communication. Benefitting from deep learning, many researchers have proposed various unsupervised models to extract effective emotional features and supervised models to train emotion recognition systems. In this paper, we utilize semi-supervised ladder networks for speech emotion recognition. The model is trained by minimizing the supervised loss and auxiliary unsupervised cost function. The addition of the unsupervised auxiliary task provides powerful discriminative representations of the input features, and is also regarded as the regularization of the emotional supervised task. We also compare the ladder network with other classical autoencoder structures. The experiments were conducted on the interactive emotional dyadic motion capture (IEMOCAP) database, and the results reveal that the proposed methods achieve superior performance with a small number of labelled data and achieves better performance than other methods.
文摘The theory of RC uniform ladder networks based upon the recurrence of voltage and cur-rent functions is extended as a vehicle to analyse the dynamic characteristics of reg lines. Meth-ods for computing the time constants and simplifying the transfer functions for reg lines are alsopresented.
基金Sponsored by the Natural Science Foundation of Heilongjiang Province (Grant No. QC2012C081)the Creative Qualified Scientists and Technicians Foundation of Harbin City (Grant No. RC2012QN001025)the National Natural Science Foundation of China (Grant No. 61107069 and 41174161)
文摘In order to improve the multiplexing capability of the optical sensors based on the lower interferential optic fiber sensing technology and the white light fiber-optic Mach-Zehnder interferometer,reflective ladder topology network ( RLT) with tailored formula was proposed. The topology network consists of 6 rungs sensing elements linked by 5 couplers. Two cases with different choices of couplers were contrasted: one is equal coupling ratio,and the other is tailored coupling ratio. Through the simulation of these two cases,the detailed multiplexing capability was analyzed,and accordingly the experiments were also carried out. The simulation results showed that,the tailored formula enhances the multiplexing capability of the structure. In the first case, the maximum number of sensors which can be multiplexed is 8,and in the other case is 12 fiber optic sensors. The experimental results have a good agreement with numerical simulation results. Thus,it is considered expedient to incorporate RLT into large-scale building,grounds,bridges,dams,tunnels,highways and perimeter security.
基金This work was supported by the Science Foundation of Fujian Provincial Key Laboratory of Polymer Materials
文摘The title complex, {[Cu2（4,4＇-bipyridine）2（μ-O2CMe）2（O2CMe）2],H2O}n 1, was synthesized and structurally characterized by X-ray crystallography. It crystallizes in monoclinic, space group C2/c with a = 13.4474（5）, b = 11.7566（2）, c = 19.5380（6）A, β = 92.930（2）°, V = 3084.84（16） A^3, Z = 4, Cu2C28N409H30, Mr = 693.64, Dc = 1.494 g/cm^3, F（000） = 1424 and μ（MoKα） = 1.436 mm^-1. With the use of 2062 observed reflections （I 〉 2σ（I））, the structure was refined to R = 0.0769 and wR = 0.2154. In complex 1, the dimeric copper acetate units are linked through 4,4’-bipyridine to yield ID molecular ladders. These ladders are connected via O-H…O hydrogen bonds to generate 2D layers, which are further linked through C-H…O hydrogen bonds to give a 3D supramolecular network.
