ABSTRACT
This paper proposes an audio data augmentation method based on deep learning in order to improve the performance of dereverberation. Conventionally, audio data are augmented using a room impulse response, which is artificially generated by some methods, such as the image method. The proposed method estimates a reverberation environment model based on a deep neural network that is trained by using clean and recorded audio data as inputs and outputs, respectively. Then, a large amount of a real augmented database is constructed by using the trained reverberation model, and the dereverberation model is trained with the augmented database. The performance of the augmentation model was verified by a log spectral distance and mean square error between the real augmented data and the recorded data. In addition, according to dereverberation experiments, the proposed method showed improved performance compared with the conventional method.
Subject(s)
Deep Learning , Databases, Factual , Models, Theoretical , Neural Networks, ComputerABSTRACT
This study presents an antibody-conjugated polydiacetylene (PDA) and its coated polyvinylidene difluoride (PVDF) membrane. The M149 antibody was hybridized to nano-vesicles consisting of pentacosa-10,12-diynoic acid (PCDA) and dimyristoylphosphatidylcholine (DMPC). After photo-polymerization at 254 nm, the effects on the PDA by antigenic injection were investigated with UV-vis spectroscopy, fluorescence spectroscopy, dynamic light scattering and transmission electron microscopy. Because PDA, an alternating ene-yne molecule, induces a blue-to-red color transition and an interesting fluorescent response by the distortion of its backbone, the biomolecular recognition of an antibodyâ»antigen can be converted into an optical and fluorescent signal. Thus, an influenza antigen was successfully detected with the proposed label-free method. Furthermore, the vesicular system was improved by coating it onto a membrane type sensing platform for its stability and portability. The proposed antibody-PDA composite PVDF membrane has potential for rapid, easy and selective visualization of the influenza virus.
ABSTRACT
Amphiphilic cyclodextrins have been synthesized with self-acylating reaction using vinyl esters in dimethylformamide. In the present study no base, catalyst, or enzyme was used, and the structural analyses using thin layer chromatography, nuclear magnetic resonance spectroscopy and mass spectrometry show that the cyclodextrin is substituted preferentially by one acyl moiety at the C2 position of the glucose unit, suggesting that cyclodextrin functions as a regioselective catalytic carbohydrate in organic solvent. In the self-acylation, the most acidic OH group at the 2-position and the inclusion complexing ability of cyclodextrin were considered to be significant. The substrate preference was also observed in favor of the long-chain acyl group, which could be attributed to the inclusion ability of cyclodextrin cavity. Furthermore, using the model amphiphilic building block, 2-O-mono-lauryl ß-cyclodextrin, the self-organized supramolecular architecture with nano-vesicular morphology in water was investigated by fluorescence spectroscopy, dynamic light scattering and transmission electron microscopy. The cavity-type nano-assembled vesicle and the novel synthetic methods for the preparation of mono-acylated cyclodextrin should be of great interest with regard to drug/gene delivery systems, functional surfactants, and carbohydrate derivatization methods.
ABSTRACT
Flavonoids are a group of plant secondary metabolites including polyphenolic molecules, and they are well known for antioxidant, anti-allergic, anti-inflammatory and anti-viral propertied. In general, flavonoids are detected with various non-colorimetric detection methods such as column liquid chromatography, thin-layer chromatography, and electrochemical analysis. For the first time, we developed a straightforward colorimetric detection system allowing recognition of some highly hydrophobic flavonoids such as alpha-naphthoflavone and beta-naphthoflavone, visually using 10,12-pentacosadiynoic acid (PCDA) derivatized with succinoglycan monomers isolated from Sinorhizobium meliloti. Besides changes in visible spectrum, we also demonstrate fluorescence changes using our detection system in the presence of those flavonoids. The succinoglycan monomers attached to PCDA molecules may function as an unstructured molecular capturer for some highly hydrophobic flavonoids by hydrophobic interactions, and transmit their molecular interactions as a color change throughout the PCDA liposome.
