In this paper, we present a comparative analysis of artificial neural networks (ANNs) and Gaussian mixture models (GMMs) for design of voice conversion system using line spectral frequencies (LSFs) as feature vectors. Both the ANN and GMM based models are explored to capture nonlinear mapping functions for modifying the vocal tract characteristics of a source speaker according to a desired target speaker. The LSFs are used to represent the vocal tract transfer function of a particular speaker. Mapping of the intonation patterns (pitch contour) is carried out using a codebook based model at segmental level. The energy profile of the signal is modified using a fixed scaling factor defined between the source and target speakers at the segmental level. Two different methods for residual modification such as residual copying and residual selection methods are used to generate the target residual signal. The performance of ANN and GMM based voice conversion (VC) system are conducted using subjective and objective measures. The results indicate that the proposed ANN-based model using LSFs feature set may be used as an alternative to state-of-the-art GMM-based models used to design a voice conversion system. 相似文献
In recent years, there has been a considerable growth of application of group technology in cellular manufacturing. This has led to investigation of the primary cell formation problem (CFP), both in classical and soft-computing domain. Compared to more well-known and analytical techniques like mathematical programming which have been used rigorously to solve CFPs, heuristic approaches have yet gained the same level of acceptance. In the last decade we have seen some fruitful attempts to use evolutionary techniques like genetic algorithm (GA) and Ant Colony Optimization to find solutions of the CFP. The primary aim of this study is to investigate the applicability of a fine grain variant of the predator-prey GA (PPGA) in CFPs. The algorithm has been adapted to emphasize local selection strategy and to maintain a reasonable balance between prey and predator population, while avoiding premature convergence. The results show that the algorithm is competitive in identifying machine-part clusters from the initial CFP matrix with significantly less number of iterations. The algorithm scaled efficiently for large size problems with competitive performance. Optimal cluster identification is then followed by removal of the bottleneck elements to give a final solution with minimum inter-cluster transition cost. The results give considerable impetus to study similar NP-complete combinatorial problems using fine-grain GAs in future. 相似文献
Multilayered wire‐grid polarizers (WGP) find application as low‐reflection polarizers in projection‐type liquid crystal display devices. A multilayered WGP is formed by adding thin layers on top of the metal ridges of an ordinary WGP. The ordinary WGP consists of a periodic array of parallel metal ridges, where the period of the array and the width of any individual metal ridge are typically less than the wavelength of the incident light. Such WGPs are often used as efficient polarizers. However, in certain applications, it is important to reduce the reflection from the WGP while preserving the polarization efficiency. One of the ways to achieve this goal is to add thin layers on top of the metal ridges of the ordinary WGP. The reduction in reflection from the multilayered WGP depends on the number and material of these additional layers. In this paper, we describe a design method for multilayered WGPs based on an effective medium theory, thin‐film computation method and a monochromatic recursive convolution finite‐difference time‐domain algorithm. The goal of design process is to identify suitable materials and thicknesses for the additional thin layers needed to lower the reflection appreciably. The design method is explained with the help of bilayered WGPs. 相似文献
Samples of three Indian coals, of widely differing origin and rank, were subjected to flash pyrolysis at a temperature of about 1150 °C for 30 s in vacuo, and under atmospheres of nitrogen, argon, ammonia, and perdeuterobenzene. The gaseous products of the pyrolyses were analysed by infra-red and mass spectroscopy and by gas chromatography. Observed variations in gas compositions are discussed relative to the possible mode of influence by the pyrolytic atmospheres. It would appear that the pyrolytic atmosphere is an important factor in determining the composition of the pyrolysis products; the influence of nitrogen, argon and perdeuterobenzene is a physical one, leading especially to higher yields of olefins. 相似文献
Over the last few years, there has been a rapid growth in digital data. Images with quotes are spreading virally through online social media platforms. Misquotes found online often spread like a forest fire through social media, which highlights the lack of responsibility of the web users when circulating poorly cited quotes. Thus, it is important to authenticate the content contained in the images being circulated online. So, there is a need to retrieve the information within such textual images to verify quotes before its usage in order to differentiate a fake or misquote from an authentic one. Optical Character Recognition (OCR) is used in this paper, for converting textual images into readable text format, but none of the OCR tools are perfect in extracting information from the images accurately. In this paper, a method of post-processing on the retrieved text to improve the accuracy of the detected text from images has been proposed. Google Cloud Vision has been used for recognizing text from images. It has also been observed that using post-processing on the extracted text improved the accuracy of text recognition by 3.5% approximately. A web-based text similarity approach (URLs and domain name) has been used to examine the authenticity of the content of the quoted images. Approximately, 96.26% accuracy has been achieved in classifying quoted images as verified or misquoted. Also, a ground truth dataset of authentic site names has been created. In this research, images with quotes by famous celebrities and global leaders have been used. A comparative analysis has been performed to show the effectiveness of our proposed algorithm.