关于汉语/英语AMR语音编码参数统计特性的研究

自适应多速率（AMR）算法是第三代移动通信系统中语音业务所采用的语音编码标准。为研究AMR算法应用于汉语语音和英语语音所产生的差异，本文针对ARM语音编码算法的特点，在处理分析了大量汉语语音数据的基础上，对算法中的线谱频率（LSF）、基音周期等关键编码参数进行了比较并给出相应结合。本文的研究有助于AMR算法针对汉语语音的性能优化。     

一种新的快速视频图像边缘检测算法

一般常用的边缘检测算法计算量较大，对细小边缘检测效果较差，不适合实时边缘检测系统。针对这一问题，本文提出一种连续分割的快速边缘检测算法：从相互垂直两方向分割梯度图像，提取截面曲线极大值获得图像边缘，使用形态学梯度，检测出细小的图像边缘。实验结果表明此算法较Canny等经典算法减少了计算量，提高了边缘检测精度。     

汉语环境的英语语音教学研究——基于汉语音素对英语语音的影响

从语言学的分支,语音学的角度出发,论述了中国英语学习者系统学习语音的重要性。中国学生在英语语音的学习中或多或少存在着受汉语影响的问题,旨在研究汉语音素对英语发音正面和反面的影响,从而提出在语音课堂教学中切实可行的方法策略。     

IP电话系统语音终端的设计及实现

以开发ＩＰ电话系统为背景，论述了语音终端及其具有实现方法，包括了对语音的实时采样、播放、自适应压缩和解压、在网络的实时传输等关键技术。同时对ＲＡＤｖｉｓｉｏｎＨ．３２３Ｓｔａｃｋ进行了一定的技术讨论，并对该系统今后的升级提出建议。     

多用户远程语音交互终端接入设备设计与实现

基于低空飞行航空管制组网话音通信应用背景,从音频接口、模数/数模转换接口和网络接口3个核心处理模块出发,研究了网络化语音接入设备的硬件关键技术。从C语言级和汇编级两方面对G.729A语音编解码算法进行改进优化,并给出了语音数据的处理流程,以单片TMS320C6455DSP实时实现了语音4路压缩编码和20路解码处理。采用TCP/IP网络协议实现了多路语音网络通信功能,分析了多路语音串扰问题并设计了相应的解决方案。测试表明,基于该技术实现的设备语音处理延迟小于10 ms,满足多用户语音实时传输要求。     

应急多方话务通信方案设计

针对突发地质灾难中无法利用常规通信网络通信的问题,提出一种基于于嵌入式微处 理器OMAP5912 的实时多用户通信方案。采用WiFi无线网卡组建Ad Hoc网络,通过AIC23芯 片实现语音采集和输出,利用动态多用户语音解码算法实现实时多用户话音通话功能。测试 表明,该系统支持多用户同时语音通信功能,语音延迟约60 ms,适合于各种应急通信 等无线应用场合。     

日本留学生汉语双音节词声调偏误调查报告

声调是留学生汉语语音学习中的一大难点。本文以日本留学生为对象,以汉语双音节词的声调为内容进行了一项调查。根据对调查结果的统计与分析归纳出日本留学生汉语双音节词声调偏误的主要类型。     

改进的基于长时谱能量差异和基音比例的语音检测方法

语音检测是语音信号处理的前端,利用长时谱能量差异特征的语音检测无法区分突 发噪声和语音,掺杂着突发噪声的语音信号会对语音处理系统带来不良影响。提出了一 种基于长时谱能量差异特征和基音比例特征相结合的语音检测方法,该方法的优点是,在利 用长时谱能量差异特征基础上引入基音比例特征,从而有效减少了将信号中突发噪声误判为 语音的错误。实验显示,该算法能够在多种信噪比环境下取得很好的检测结果。     

ITU G.732.1语音编码器在DSP上的实时实现

ＩＴＵＧ。７２３．１建议的编码速度为５．３和６．３ｂｉｔ／ｓ的双码率编码方案是一种运算复杂的用于多媒体通信中的低码率语音编码器。本文介绍了该编码器的算法原理和在一片ＴＭＳ３２０Ｃ５４１定点ＤＳＰ芯片上实时实现该编码器过程中的软，硬件结构及关键技术。     

数字化语音减速保调播放插值算法及实现

对数字化语音减速保调播放问题进行了研究并提出了一种播值算法。实验表明，算法既保证了语音质量，又有良好的实时性。     

一种基于低比特率语音算法的芯片设计与分析

介绍了一种低比特率语音算法,它是在MBE编码算法的基础上,利用线性预测谱代替了MBE中的傅里叶变换谱对MBE编码算法进行了改进,从而在保持语音质量的情况下使比特率更低,达到2 kbit/s.同时阐述了基于该算法的系统芯片设计,并对该算法中的基音周期参数、V/U等参数的提取进行详细的推导分析.最后对本算法与常用的码激励线性预测编码(CELP)的算法进行比较,并分析其中的原因.     

Spending time,spending money: passenger segmentation in an international airport

Changes within the air transport sector have required many European airports to either develop or expand their commercial activities. Strategies have included the expansion of retail space, a broadening of the tenant and merchandise mix and the development of a passenger segmentation strategy. This paper explores the efficacy of this approach by identifying the behaviour of different passenger segments while in an international airport. Using a framework of strategic market segmentation, it identifies how travellers allocate their time having entered ‘airside’ and details any purchases made. Using observational research and a face-to-face quantitative survey, 301 passengers were tracked and interviewed. Through a broad based, a priori form of segmentation, significant differences in shopping behaviour are identified. Such findings assist with the development of the airport's commercial strategy and allow a number of observations to be made about the value of market segmentation from both a theoretical and managerial perspective.     

Corporate Rights to Free Speech?

Although the courts have ruled that companies are legal persons, they have not yet made clear the extent to which political free speech for corporations is limited by the strictures legitimately placed upon corporate commercial speech. I explore the question of whether or not companies can properly be said to have the right to civil free speech or whether corporate speech is always de facto commercial speech not subject to the same sorts of legal protections as is the right to civil free speech. In the absence of clearly defined legal precedent, I emphasize moral reasons for determining the appropriate limits of corporate civil free speech. Appealing to arguments typically used to justify individual rights to civil free speech, I examine the extent to which this sort of justification may or may not be legitimately extended to corporations. I conclude that corporate rights to civil free speech must be restricted because granting rights of free speech to institutions may, in practice, undermine the moral rationale and practical feasibility of guaranteeing rights of civil free speech to individuals. Furthermore, I argue that granting corporations full rights to civil free speech will undercut attempts to develop good moral character in corporate institutions by undermining the efforts of watchdog organizations.     

Speed-profile-based road segmentation for accident occurrence modelling for hilly terrains

Modelling the accident occurrences per unit length of the road requires segmentation of road stretch. In the case of hilly terrains, homogeneity-based approach results in extremely small sections thus creating a biased accident data. This result is due to the high variations in geometry along the length of the road stretch. Constant length sections do not consider the effect of road's characteristics, hence may lead to undesirable results. It is hypothesized that the speed profile of a vehicle moving in free-flow conditions is governed by the road geometry as well as the road side environment. In this paper, a new methodology based on speed profile of a test vehicle (passenger car) has been proposed for segmenting a road stretch passing through hilly terrain. Comparative analysis of the results obtained from the constant length segmentation approach and the proposed approach shows that the later approach is producing statistically better results. Relatively more parameters were found to be statistically significant in the model based on speed-profile-based segmentation. Grade length, which could be one of the major parameters in hilly terrains, was found to be statistically significant only in the model based on speed-profile-based segmentation.     

Optimal Duration of Promotion for Durable Technology Product in a Segmented Market

ABSTRACT

Firms develop segment driven promotional mix to target its potential customers selectively. As the marginal revenues from promotion decreases with time, it is important to determine the optimal stopping time of a promotional activity. In the literature, the studies carried in this domain do not consider market segmentation and assume a uniform promotional mix. In this study, alternative decision models are developed to determine the duration of promotion for a durable technology product applicable to a segmented market. A case study is discussed to validate the proposed models and to illustrate the solution methodology based on a differential evolution algorithm.     

一种雷达回波数据的实时压缩算法

本文研究了脱靶量测量中的雷达回波数据的实时压缩问题，通过对作匀速直线运动目标的多普勒信号特征的分析，提出了一种基于离散余弦变换（DCT）的多普勒信号的变换域压缩算法，并讨论了根据重构信号由高分辨率谱估计及非线性最小二乘拟合进行脱靶量测量的问题。     

一种基于Levy-人工蜂群的三维Otsu阈值分割算法

针对图像分割过程中三维Otsu算法运算时间长、计算量大的问题,提出了一种基于Levy-人工蜂群算法的三维Otsu阈值分割算法。首先,以像素灰度值-邻域均值-邻域中值的三维类间方差作为人工蜂群算法的适应度函数;其次,采用Levy飞行模式评价像素的适应度,对其种群更新及邻域搜索过程进行优化,以增强其全局搜索能力;最后,利用改进后的算法得到的分割阈值对图像进行分割。仿真实验结果表明,与传统三维Otsu阈值分割算法相比,所提算法能够有效降低图像存储空间,处理时间降低了30.8%,具备更好的抗噪性能,分割效果也更为理想。     

融合改进FT显著性与Grabcut的图像目标分割算法

针对目前复杂度较大的图像中目标分割速度较慢、显著性边界分割不明确等问题,提出了一种融合改进的FT（Frequency-tuned）显著性检测与Grabcut的图像分割算法。该算法首先通过改进基于频率调谐的FT显著性检测方法得到图像中显著性较高的区域,并利用SLIC（Simple Linear Iterative Clustering）算法对显著图进行预处理得到超像素图,能够有效改善边界的分割效果,然后通过以图论GraphCut算法为基础改进的Grabcut算法建立高斯混合模型。为了提高算法效率,通过聚类以超像素代替原像素,并反复迭代高斯混合模型（Gaussian Mixed Model,GMM）参数,最后利用最大流最小割算法得到最优目标分割结果。实验结果表明所提算法能够更准确更高效率地分割图像中的显著性目标,对高分辨率图像也有很好的适用效果,相比于其他算法在分割精度上提高10%左右,并具有较高的分割效率。     

A Cluster Analytic Approach for Consumer Segmentation Using the Vegetarian/Meatarian Distinction

Abstract

In this study, we use cluster analysis to create a consumer segmentation scheme based on individuals' vegetarian/meatarian orientation, along with key attitudinal, personality, and demographic characteristics. Such characteristics include concern for animals, concern with health, concern for the environment, attitude toward corporate ethical practices, need for cognition, social conformity, age, gender, and level of education. In segmenting consumers along these dimensions, this research study not only expands our knowledge about consumers, but also provides useful insight for marketers of food products and services.     

Culturally Customizing Websites for Immigrant Communities in the United States

Recent research suggests that the customization of online marketing messages to meet the cultural preferences of immigrant communities increases preferences for those messages. Level of acculturation may be used as a segmentation variable that accounts for this effect, with lower levels of acculturation matching the preference for the home country's culture in marketing messages. Comparisons between Asian- and Hispanic-American preferences across levels of acculturation are made, and possible convergences in preferences through acculturation are examined. The results indicate that Taiwanese-Americans have culturally-rooted preferences for web content based on their acculturation levels. Furthermore, the study shows that the web content preferences of Taiwanese- and Hispanic-American consumers converge as members of these groups acculturate.