The recently proposed random-phase-free method enables holographic reconstructions with very low noise,which allows fine projections without time integration of sub-holograms. Here, we describe the additional advantag...The recently proposed random-phase-free method enables holographic reconstructions with very low noise,which allows fine projections without time integration of sub-holograms. Here, we describe the additional advantage of this method, namely, the extended depth of sharp imaging. It can be attributed to a lower effective aperture of the hologram section forming a given image point at the projection screen. We experimentally compare the depth of focus and imaging resolution for various defocusing parameters in the cases of the random-phase method and the random-phase-free method. Moreover, we discuss the influence of the effective aperture in the presence of local obstacles in the hologram＇s plane.展开更多
For text-independent speaker verification, the Gaussian mixture model （GMM） using a universal background model strategy and the GMM using support vector machines are the two most commonly used methodologies. Recentl...For text-independent speaker verification, the Gaussian mixture model （GMM） using a universal background model strategy and the GMM using support vector machines are the two most commonly used methodologies. Recently, a new SVM-based speaker verification method using GMM super vectors has been proposed. This paper describes the construction of a new speaker verification system and investigates the use of nuisance attribute projection and test normalization to further enhance performance. Experiments were conducted on the core test of the 2006 NIST speaker recognition evaluation corpus. The experimental results indicate that an SVM-based speaker verification system using GMM super vectors can achieve appealing performance. With the use of nuisance attribute projection and test normalization, the system performance can be significantly improved, with improvements in the equal error rate from 7.78% to 4.92% and detection cost function from 0.0376 to 0.0251.展开更多
基金funded by the Polish National Centre for Research and Development(No.LIDER/013/469/L-4/12/NCBR/2013)the Polish National Science Centre(No.2015/17/B/ST7/03754)
文摘The recently proposed random-phase-free method enables holographic reconstructions with very low noise,which allows fine projections without time integration of sub-holograms. Here, we describe the additional advantage of this method, namely, the extended depth of sharp imaging. It can be attributed to a lower effective aperture of the hologram section forming a given image point at the projection screen. We experimentally compare the depth of focus and imaging resolution for various defocusing parameters in the cases of the random-phase method and the random-phase-free method. Moreover, we discuss the influence of the effective aperture in the presence of local obstacles in the hologram＇s plane.
文摘For text-independent speaker verification, the Gaussian mixture model （GMM） using a universal background model strategy and the GMM using support vector machines are the two most commonly used methodologies. Recently, a new SVM-based speaker verification method using GMM super vectors has been proposed. This paper describes the construction of a new speaker verification system and investigates the use of nuisance attribute projection and test normalization to further enhance performance. Experiments were conducted on the core test of the 2006 NIST speaker recognition evaluation corpus. The experimental results indicate that an SVM-based speaker verification system using GMM super vectors can achieve appealing performance. With the use of nuisance attribute projection and test normalization, the system performance can be significantly improved, with improvements in the equal error rate from 7.78% to 4.92% and detection cost function from 0.0376 to 0.0251.
