In this paper, we propose two kinds of modifications in speaker recognition. First, the correlations between frequency channels are of prime importance for speaker recognition. Some of these correlations are lost when...In this paper, we propose two kinds of modifications in speaker recognition. First, the correlations between frequency channels are of prime importance for speaker recognition. Some of these correlations are lost when the frequency domain is divided into sub-bands. Consequently we propose a particularly redundant parallel architecture for which most of the correlations are kept. Second, generally a log transformation used to modify the power spectrum is done after the filter-bank in the classical spectrum calculation. We will see that performing this transformation before the filter bank is more interesting in our case. In the processing of recognition, the Gaussian mixture model (GMM) recognition arithmetic is adopted. Experiments on speech corrupted by noise show a better adaptability of this approach in noisy environments, comoared with a conventional device, esoeciallv when oruning of some recognizers is performed.展开更多
Objective:To examine whether exposure of mouse neuronal cells to radiofrequency fields used in mobile communication devices can induce stress in endoplasmic reticulum(ER)and activate unfolded protein response(UPR).Met...Objective:To examine whether exposure of mouse neuronal cells to radiofrequency fields used in mobile communication devices can induce stress in endoplasmic reticulum(ER)and activate unfolded protein response(UPR).Methods:HT22 mouse hippocampus neuronal cells were exposed to continuous wave 900 MHz radiofrequency fields(RF)at 120μW/cm2 power intensity for 4 h/d for 5 consecutive days.The positive control cells were irradiated with 4 Gy of 60Coγ-rays at a dose rate of 0.5 Gy/min(GR).Twenty-four hours after the last exposure,cells were collected,and the expressions of sensor transmembrane proteins were detected using Western blot analysis.Results:The expression levels of Ire1,PERK,p-IRE1 and p-PERK,GRP78 and CHOP proteins were detected.There were no statistically significant differences in the expression levels of IRE1 and PERK proteins in control(CT),sham(SH)-,RF-and GR-exposed cells(P<0.05).The phosphorylated protein levels of p-IRE1 and p-PERK were significantly increased in cells exposed to RF and GR(P<0.05).The expression levels of GRP78 and CHOP were significantly increased in RF-and GR-exposed cells compared to CT and SH-exposed cells(P<0.05).Cells treated with 1μg/ml TM for 24 h showed significantly increased expression levels of GRP78 and CHOP proteins compared to controls(P<0.05).In the presence of 2 mmol/L PBA,TM-induced increased levels of GRP78 and CHOP proteins were reduced(P<0.05).Conclusions:The exposure of non-ionizing 900 MHz RF was able to cause stress in HT22 mouse neuronal cells and activated UPR in ER.Since UPR plays an important role in both cell survival(when UPR is mitigated)and apoptosis/death(under unresolvable stress conditions),further studies are required to determine the fate of the cells exposed to RF.展开更多
基金the National Natural Science Foundation of China (No. 60171043, 60371046)
文摘In this paper, we propose two kinds of modifications in speaker recognition. First, the correlations between frequency channels are of prime importance for speaker recognition. Some of these correlations are lost when the frequency domain is divided into sub-bands. Consequently we propose a particularly redundant parallel architecture for which most of the correlations are kept. Second, generally a log transformation used to modify the power spectrum is done after the filter-bank in the classical spectrum calculation. We will see that performing this transformation before the filter bank is more interesting in our case. In the processing of recognition, the Gaussian mixture model (GMM) recognition arithmetic is adopted. Experiments on speech corrupted by noise show a better adaptability of this approach in noisy environments, comoared with a conventional device, esoeciallv when oruning of some recognizers is performed.
基金This research is supported by funding from the National Natural Science Foundation of China(Grant No.81373025).
文摘Objective:To examine whether exposure of mouse neuronal cells to radiofrequency fields used in mobile communication devices can induce stress in endoplasmic reticulum(ER)and activate unfolded protein response(UPR).Methods:HT22 mouse hippocampus neuronal cells were exposed to continuous wave 900 MHz radiofrequency fields(RF)at 120μW/cm2 power intensity for 4 h/d for 5 consecutive days.The positive control cells were irradiated with 4 Gy of 60Coγ-rays at a dose rate of 0.5 Gy/min(GR).Twenty-four hours after the last exposure,cells were collected,and the expressions of sensor transmembrane proteins were detected using Western blot analysis.Results:The expression levels of Ire1,PERK,p-IRE1 and p-PERK,GRP78 and CHOP proteins were detected.There were no statistically significant differences in the expression levels of IRE1 and PERK proteins in control(CT),sham(SH)-,RF-and GR-exposed cells(P<0.05).The phosphorylated protein levels of p-IRE1 and p-PERK were significantly increased in cells exposed to RF and GR(P<0.05).The expression levels of GRP78 and CHOP were significantly increased in RF-and GR-exposed cells compared to CT and SH-exposed cells(P<0.05).Cells treated with 1μg/ml TM for 24 h showed significantly increased expression levels of GRP78 and CHOP proteins compared to controls(P<0.05).In the presence of 2 mmol/L PBA,TM-induced increased levels of GRP78 and CHOP proteins were reduced(P<0.05).Conclusions:The exposure of non-ionizing 900 MHz RF was able to cause stress in HT22 mouse neuronal cells and activated UPR in ER.Since UPR plays an important role in both cell survival(when UPR is mitigated)and apoptosis/death(under unresolvable stress conditions),further studies are required to determine the fate of the cells exposed to RF.