Study on Closed-set Speaker Identification Based on Biomimetic Pattern Recognition

In this paper, a new classifier of speaker identification has been proposed, which is based on Biomimetic pattern recognition (BPR). Distinguished from traditional speaker recognition methods, such as DWT, HMM, GMM, SVM and so on, the proposed classifier is constructed by some finite sub-space which is reasonable covering of the points in high dimensional space according to distributing characteristic of speech feature points. It has been used in the system of speaker identification. Experiment results show that better effect could be obtained especially with lesser samples. Furthermore, the proposed classifier employs a much simpler modeling structure as compared to the GMM. In addition, the basic idea "cognition" of Biomimetic pattern recognition (BPR) results in no requirement of retraining the old system for enrolling new speakers.

Information geometry theory of high-dimension space and application for speaker independent continuous digit speech recognition

In the light of descriptive geometry and notions in set theory, this paper re-defines the basic elements in space such as curve and surface and so on, presents some fundamental notions with respect to the point cover based on the High-dimension space (HDS) point covering theory, finally takes points from mapping part of speech signals to HDS, so as to analyze distribution information of these speech points in HDS, and various geometric covering objects for speech points and their relationship. Besides, this paper also proposes a new algorithm for speaker independent continuous digit speech recognition based on the HDS point dynamic searching theory without end-points detection and segmentation. First from the different digit syllables in real continuous digit speech, we establish the covering area in feature space for continuous speech. During recognition, we make use of the point covering dynamic searching theory in HDS to do recognition, and then get the satisfying recognized results. At last, compared to HMM (Hidden Markov models)-based method, from the development trend of the comparing results, as sample amount increasing, the difference of recognition rate between two methods will decrease slowly, while sample amount approaching to be very large, two recognition rates all close to 100% little by little. As seen from the results, the recognition rate of HDS point covering method is higher than that of in HMM (Hidden Markov models) based method, because, the point covering describes the morphological distribution for speech in HDS, whereas HMM-based method is only a probability distribution, whose accuracy is certainly inferior to point covering.

Study on text-dependent speaker recognition based on Biomimetic Pattern Recognition

We studied the application of Biomimetic Pattern Recognition to speaker recognition. A speaker recognition neural network using network matching degree as criterion is proposed. It has been used in the system of text-dependent speaker recognition. Experimental results show that good effect could be obtained even with lesser samples. Furthermore, the misrecognition caused by untrained speakers occurring in testing could be controlled effectively. In addition, the basic idea "cognition" of Biomimetic Pattern Recognition results in no requirement of retraining the old system for enrolling new speakers.

A speaker-independent continuous speech recognition system using biomimetic pattern recognition

In speaker-independent speech recognition, the disadvantage of the most diffused technology (HMMs, or Hidden Markov models) is not only the need of many more training samples, but also long train time requirement. This paper describes the use of Biomimetic pattern recognition (BPR) in recognizing some mandarin continuous speech in a speaker-independent manner. A speech database was developed for the course of study. The vocabulary of the database consists of 15 Chinese dish's names, the length of each name is 4 Chinese words. Neural networks (NNs) based on Multi-weight neuron (MWN) model are used to train and recognize the speech sounds. The number of MWN was investigated to achieve the optimal performance of the NNs-based BPR. This system, which is based on BPR and can carry out real time recognition reaches a recognition rate of 98.14% for the first option and 99.81% for the first two options to the persons from different provinces of China speaking common Chinese speech. Experiments were also carried on to evaluate Continuous density hidden Markov models (CDHMM), Dynamic time warping (DTW) and BPR for speech recognition. The Experiment results show that BPR outperforms CDHMM and DTW especially in the cases of samples of a finite size.

Biomimetic pattern recognition for speaker-independent speech recognition

In speaker-independent speech recognition, the disadvantage of the most diffused technology ( Hidden Markov Models) is not only the need of many more training samples, but also long train time requirement. This paper describes the use of Biomimetic Pattern Recognition (BPR) in recognizing some Mandarin Speech in a speaker-independent manner. The vocabulary of the system consists of 15 Chinese dish's names. Neural networks based on Multi-Weight Neuron (MWN) model are used to train and recognize the speech sounds. Experimental results are presented to show that the system, which can carry out real time recognition of the persons from different provinces speaking common Chinese speech, outperforms HMMs especially in the cases of samples of a finite size.

Study on text-dependent speaker recognition based on Biomimetic Pattern Recognition

We studied the application of Biomimetic Pattern Recognition to speaker recognition. A speaker recognition neural network using network matching degree as criterion is proposed. It has been used in the system of text-dependent speaker recognition. Experimental results show that good effect could be obtained even with lesser samples. Furthermore, the misrecognition caused by untrained speakers occurring in testing could be controlled effectively. In addition, the basic idea "cognition" of Biomimetic Pattern Recognition results in no requirement of retraining the old system for enrolling new speakers.

Formation of low surface energy separators with undercut structures via a full-solution process and its application in inkjet printed matrix of polymer light-emitting diodes

In this paper, low surface energy separators With undercut structures were fabricated through a full solution process, These low Surface energy separators are more suitable for application in inkjet printed passive-matrix displays of polymer light-emitting diodes. A patterned PS film was formed on the P4VP/photoresist film by microtransfer printing firstly. Patterned Au-coated Ni film was formed on the uncovered P4VP/photoresist film by electroless deposition. This metal film was used as mask to pattern the photoresist layer and form undercut structures with the patterned photoresist layer. The surface energy of the metal film also decreased dramatically from 84.6 mj/m(2) to 21.1 mJ/m(2) by modification of fluorinated mercaptan self-assemble monolayer on Au surface. The low surface energy separators were used to confine the flow of inkjet printed PFO solution and improve the patterning resolution of inkjet printing successfully. Separated PFO stripes, complement with the pattern of the separators, formed through inkjet printing.

Antioxidant Sensors Based on Iron Diethylenetriaminepentaacetic Acid, Hematin, and Hemoglobin Modified TiO2 Nanoparticle Printed Electrodes

Antioxidant amperometric sensors based on iron-containing complexes and protein modified electrodes were developed. Indium tin oxide glass was printed with TiO2 nanoparticles, onto which iron-containing compounds and protein were adsorbed. When applied with negative potentials, the dissolved oxygen is reduced to H2O2 at the electrode surface, and the H2O2 generated in situ oxidizes Fe-II to Fe-III, and then electrochemical reduction of Fe-III therefore gives rise to a catalytic current. In the presence of antioxidants, H2O2 was scavenged, the catalytic current was reduced, and the decreased current signal was proportional to the quantity of existing antioxidants. A kinetic model was proposed to quantify the H2O2 scavenging capacities of the antioxidants. With the use of the sensor developed here, antioxidant measurements can be done quite simply: put the sensor into the sample solutions (in aerobic atmosphere), perform a cathodic polarization scan, and then read the antioxidant activity values. The present work can be complementary to the previous studies of antioxidant sensor techniques based on OH radicals and superoxide ions scavenging methods, but the sensor developed here is much easier to fabricate and use.

Layer-by-layer electrodeposition of redox polymers and enzymes on screen-printed carbon electrodes for the preparation of reagentless biosensors

Layer-by-layer electrodeposition of redox polymer/enzyme composition films on screen-printed carbon electrodes for fabrication of reagentless enzyme biosensors has been proposed and the resulting films were found to be very stable and rigid.

Electrooxidative polymerization of phenothiazine derivatives on screen-printed carbon electrode and its application to determine NADH in flow injection analysis system

The electrooxidation polymerization of phenothiazine derivatives, including azure A and toluidine blue 0, has been studied at screen-printed carbon electrodes in neutral phosphate buffer. Both compounds yield strongly adsorbed electroactive polymer with reversible behavior and formal potentials closed to 0.04 V at pH 6.9. The modified electrodes exhibited good stability and electrocatalysis for NADH oxidation in phosphate buffer (pH 6.9), with an overpotential of more than 500 mV lower than that of the bare electrodes. Further, the modified screen-printed carbon electrodes were found to be promising as an amperometric detector for the flow injection analysis (FIA) of NADH, typically with a dynamic range of 0.5-100 muM.

The modification of screen-printed carbon electrodes with amino group and its application to construct a H2O2 biosensor

Electrooxidation of thionine on screen-printed carbon electrode gives rise to the modification of the surface with amino groups for the covalent immobilization of enzymes such as horseradish peroxidase (HRP). The biosensor was constructed using multilayer enzymes which covalently immobilized onto the surface of amino groups modified screen-printed carbon electrode using glutaraldehyde as a bifunctional reagent. The multilayer assemble of HRP has been characterized with the cyclic voltammetry and the faradaic impedance spectroscopy. The H2O2 biosensor exhibited a fast response (2 s) and low detection limit (0.5 muM).

Electropolymerization of azure B on a screen-printed carbon electrode and its application to the determination of NADH in a flow injection analysis system

The electrooxidation polymerization of azure B on screen-printed carbon electrodes in neutral phosphate buffer was studied. The poly(azure B) modified electrodes exhibited excellent electrocatalysis and stability for dihydronicotinamide adenine dinucleotide (NADH) oxidation in phosphate buffer (pH 6.9), with an overpotential of more than 400 mV lower than that at the bare electrodes. Different techniques, including cyclic voltammetry, scanning electron microscopy (SEM) and X-ray photoelectron spectroscopy have been employed to characterize the poly (azure B) film. Furthermore, the modified screen-printed carbon electrodes were found to be promising as an amperometric detector for the flow injection analysis (FIA) of NADH, typically with a dynamic range of 0.5 muM to 100 muM.

Multilayer films of carbon nanotubes and redox polymer on screen-printed carbon electrodes for electrocatalysis of ascorbic acid

Multilayer films containing multiwall carbon nanotubes and redox polymer were successfully fabricated on a screen-printed carbon electrode using layer-by-layer (LBL) assembled method. UV-vis spectroscopy, X-ray photoelectron spectroscopy, field-emission scanning electron microscopy and electrochemical method were used to characterize the assembled multilayer films. The multilayer films modified electrodes exhibited good electrocatalytic activity towards the oxidation of ascorbic acid (AA). Compared with the bare electrode, the oxidation peak potential negatively shifted about 350 mV (versus Ag/AgCl). Furthermore, the modified screen-printed carbon electrodes (SPCEs) could be used for the determination of ascorbic acid in real samples.

Flow injection electrochemical enzyme immunoassay based on the use of an immunoelectrode strip integrate immunosorbent layer and a screen-printed carbon electrode

A flow injection amperometric immunoassay system based on the use of screen-printed carbon electrode for the detection of mouse IgG was developed. An immunoelectrode strip, on which an immunosorbent layer and screen-printed carbon electrode were integrated, and a proposed flow cell have been fabricated. The characterization of the flow immunoassay system and parameters affecting the performance of the immunoassay system were studied and optimized. Amperometric detection at 0.0 V (versus Ag/AgCl) resulted in a linear detection range of 30-700 ng ml(-1), with a detection limit of 3 ng ml(-1). The signal variation among electrode strips prepared from variant batch did not exceed 8.5% (n = 7) by measuring 0.5 mug ml(-1) antigen standard solution.