Nagoya Institute of Technology

About: Nagoya Institute of Technology is a education organization based out in Nagoya, Japan. It is known for research contribution in the topics: Thin film & Catalysis. The organization has 10766 authors who have published 19140 publications receiving 255696 citations. The organization is also known as: Nagoya Kōgyō Daigaku & Nitech.

An Excitation Model for HMM-Based Speech Synthesis Based on Residual Modeling

Abstract: This paper describes a trainable excitation approach to eliminate the unnaturalness of HMM-based speech synthesizers. During the waveform generation part, mixed excitation is constructed by state-dependent filtering of pulse trains and white noise sequences. In the training part, filters and pulse trains are jointly optimized through a procedure which resembles analysis-bysynthesis speech coding algorithms, where likelihood maximization of residual signals (derived from the same database which is used to train the HMM-based synthesizer) is pursued. Preliminary results show that the novel excitation model in question eliminates the unnaturalness of synthesized speech, being comparable in quality to the the best approaches thus far reported to eradicate the buzziness of HMM-based synthesizers.

Characterization of Electrode/Electrolyte Interface with X-Ray Reflectometry and Epitaxial-Film LiMn2O4 Electrode

Abstract: Structural changes at electrode/electrolyte interface of a lithium cell were studied by X-ray reflectometry and two-dimensional model electrodes with a restricted lattice plane of LiMn 2 O 4 . The electrodes were constructed with an epitaxial film synthesized by the pulsed laser deposition method. The orientation of the film depends on the substrate plane; the (111), (110), and (100) planes of LiMn 2 O 4 grew on the (111), (110), and (100) planes of the SrTiO 3 substrates, respectively. The ex situ reflectometry indicated that a thin impurity layer covered the lattice plane of the as-grown film. The impurity layer was dissolved and a solid-electrolyte-interface-like phase appeared after the electrode was soaked into the electrolyte. A defect layer was formed in the (111) plane, whereas no density changes were detected for the other lattice planes. The in situ observation clarified that the surface reactivity depended on the lattice planes of the spinel; the defect layer at the (111) plane was stable during the electrochemical reaction, whereas a slight decrease in the film thickness was observed for the (110) plane. Our surface characterization of the intercalation electrode indicated that the surface structure changes during the pristine stage of the change-discharge processes and these changes are dependent on the lattice orientation of LiMn 2 O 4 .

Electrocatalytic Hydrogen Production by a Nickel(II) Complex with a Phosphinopyridyl Ligand

Abstract: A novel nickel(II) complex [Ni(L)2Cl]Cl with a bidentate phosphinopyridyl ligand 6-((diphenylphosphino)methyl)pyridin-2-amine (L) was synthesized as a metal-complex catalyst for hydrogen production from protons. The ligand can stabilize a low Ni oxidation state and has an amine base as a proton transfer site. The X-ray structure analysis revealed a distorted square-pyramidal NiII complex with two bidentate L ligands in a trans arrangement in the equatorial plane and a chloride anion at the apex. Electrochemical measurements with the NiII complex in MeCN indicate a higher rate of hydrogen production under weak acid conditions using acetic acid as the proton source. The catalytic current increases with the stepwise addition of protons, and the turnover frequency is 8400 s−1 in 0.1 m [NBu4][ClO4]/MeCN in the presence of acetic acid (290 equiv) at an overpotential of circa 590 mV.

Incremental learning methods with retrieving of interfered patterns

Abstract: There are many cases when a neural-network-based system must memorize some new patterns incrementally. However, if the network learns the new patterns only by referring to them, it probably forgets old memorized patterns, since parameters in the network usually correlate not only to the old memories but also to the new patterns. A certain way to avoid the loss of memories is to learn the new patterns with all memorized patterns. It needs, however, a large computational power. To solve this problem, we propose incremental learning methods with retrieval of interfered patterns (ILRI). In these methods, the system employs a modified version of a resource allocating network (RAN) which is one variation of a generalized radial basis function (GRBF). In ILRI, the RAN learns new patterns with a relearning of a few number of retrieved past patterns that are interfered with the incremental learning. We construct ILRI in two steps. In the first step, we construct a system which searches the interfered patterns from past input patterns stored in a database. In the second step, we improve the first system in such a way that the system does not need the database. In this case, the system regenerates the input patterns approximately in a random manner. The simulation results show that these two systems have almost the same ability, and the generalization ability is higher than other similar systems using neural networks and k-nearest neighbors.