视听翻译模式Voice-over的研究综述

本文聚焦视听翻译模式Voice-over,分析了国内外发表于翻译研究相关学术期刊上的相关学术论文,探讨国内外Voice-over模式的具体概念、发展阶段与应用领域等问题,认为国内外翻译研究界对Voice-over这一视听翻译模式的整体关注度不高,但近年来有增强的趋势;该模式除了应用于视听翻译领域外,开始应用于其他领域;学科意识逐渐增强,呈现多学科互相促进之势。本文对Voice-over模式的发展阶段、分类研究与国内研究现状做出了整理,以期为国内该模式的研究发展有所助益。

Modular Extremely Large-Scale Array Communication:Near-Field Modelling and Performance Analysis

This paper investigates the wireless communication with a novel architecture of antenna arrays,termed modular extremely large-scale array(XLarray),where array elements of an extremely large number/size are regularly mounted on a shared platform with both horizontally and vertically interlaced modules.Each module consists of a moderate/flexible number of array elements with the inter-element distance typically in the order of the signal wavelength,while different modules are separated by the relatively large inter-module distance for convenience of practical deployment.By accurately modelling the signal amplitudes and phases,as well as projected apertures across all modular elements,we analyse the near-field signal-to-noise ratio(SNR)performance for modular XL-array communications.Based on the non-uniform spherical wave(NUSW)modelling,the closed-form SNR expression is derived in terms of key system parameters,such as the overall modular array size,distances of adjacent modules along all dimensions,and the user's three-dimensional(3D)location.In addition,with the number of modules in different dimensions increasing infinitely,the asymptotic SNR scaling laws are revealed.Furthermore,we show that our proposed near-field modelling and performance analysis include the results for existing array architectures/modelling as special cases,e.g.,the collocated XL-array architecture,the uniform plane wave(UPW)based far-field modelling,and the modular extremely large-scale uniform linear array(XL-ULA)of onedimension.Extensive simulation results are presented to validate our findings.

Voice Control for an Industrial Robot as a Combination of Various Robotic Assembly Process Models

Models for the design of assembly processes are considered. Various models for the voice control of an industrial robot are considered: a logical model, semantic networks, a frame model and Petri nets. It is shown that this set of models allows describing the process of designing the technological process for an industrial robot. The logical model of the technological process allows you to define logical relationships. A model based on semantic networks describes the relationship between assembly units in a detail. This allows you to determine the order and method of registration, as well as the mutual orientation of assembly units in the product. The frame model provides the ability to streamline the execution of the build process. A model based on Petri nets allows one to describe the type and sequence of technological transitions. Based on the proposed models, a method of voice control for an industrial robot is developed. The basic principles of voice control for an industrial robot are considered.

Impacts of Model Mismatch and Array Scale on Channel Estimation for XL-HRIS-Aided Systems

Extremely large-scale hybrid reconfigurable intelligence surface(XL-HRIS),an improved version of the RIS,can receive the incident signal and enhance communication performance.However,as the RIS size increases,the phase variations of the received signal across the whole array are nonnegligible in the near-field region,and the channel model mismatch,which will decrease the estimation accuracy,must be considered.In this paper,the lower bound(LB)of the estimated parameter is studied and the impacts of the distance and signal-tonoise ratio(SNR)on LB are then evaluated.Moreover,the impacts of the array scale on LB and spectral efficiency(SE)are also studied.Simulation results verify that even in extremely large-scale array systems with infinite SNR,channel model mismatch can still limit estimation accuracy.However,this impact decreases with increasing distance.

ON USING NON-LINEAR CANONICAL CORRELATION ANALYSIS FOR VOICE CONVERSION BASED ON GAUSSIAN MIXTURE MODEL

Voice conversion algorithm aims to provide high level of similarity to the target voice with an acceptable level of quality.The main object of this paper was to build a nonlinear relationship between the parameters for the acoustical features of source and target speaker using Non-Linear Canonical Correlation Analysis(NLCCA) based on jointed Gaussian mixture model.Speaker indi-viduality transformation was achieved mainly by altering vocal tract characteristics represented by Line Spectral Frequencies(LSF).To obtain the transformed speech which sounded more like the target voices,prosody modification is involved through residual prediction.Both objective and subjective evaluations were conducted.The experimental results demonstrated that our proposed algorithm was effective and outperformed the conventional conversion method utilized by the Minimum Mean Square Error(MMSE) estimation.

Emoti-Shing: Detecting Vishing Attacks by Learning Emotion Dynamics through Hidden Markov Models

This study examines vishing, a form of social engineering scam using voice communication to deceive individuals into revealing sensitive information or losing money. With the rise of smartphone usage, people are more susceptible to vishing attacks. The proposed Emoti-Shing model analyzes potential victims’ emotions using Hidden Markov Models to track vishing scams by examining the emotional content of phone call audio conversations. This approach aims to detect vishing scams using biological features of humans, specifically emotions, which cannot be easily masked or spoofed. Experimental results on 30 generated emotions indicate the potential for increased vishing scam detection through this approach.

QoE-driven voice quality evaluation model of VoIP based on network simulation

The quality of experience（ QoE） evaluation model for voice over IP（ VoI P） service is studied to analyze the impact of network parameters on voice quality and monitor voice quality in real-time for operators.Firstly,the influence of some network parameters on the voice quality of VoI P is investigated. Then,a simulation platform for VoI P transmission is built to collect voice data under different network enviornments. According to the simulation results,a new mapping model between these arguments and VoI P voice quality is deduced. Finally,the accuracy of this voice quality evaluation model is examined and the results demanstrate that it has high reliability and feasibility.

VoiceXML语音平台中预取方案的研究

该文在分析目前主要预取算法优劣的基础上,根据VoiceXML语音平台与基于HTML的WWW之间的区别,认为在VoiceXML语音平台中应该预取其引用的语音资源,提出一种自适应的多用户共享的Markov预测模型,统一预测所有在线用户下一步所需的资源及其访问概率,有助于提高预测的准确率。最后,该文还提出抢占式优先级模型来调度预取任务,将资源的访问概率映射为优先级。仿真研究表明,与单用户预测算法和循环调度模型比较,该预取算法和调度模型都能很好地减少用户请求的访问延迟,提高响应速度。

“voice”互动模型下学术语篇中的“声音”解读

声音是学术语篇的基本特征,是作者在学术界实现身份认可、交流的重要渠道。文章以Hyland的"voice"互动模型为基础,以学术语篇中声音构建的语言手段为切入点,分析了学术语篇中作者的声音和读者的声音内在的关系及功能。研究发现:作者的声音与读者的声音在语篇层面存在互动并相互制约,且以作者的声音为主导,进而形成学术语篇的论点。

员工建言策略的VOICE模型——基于访谈数据的探索性研究

尽管建言行为对组织创新与发展非常重要,但对建言者来说却是一项极具风险的行为。建言不但可能被忽视,还可能因建言不当激怒上级而引发负面效应。因此,建言者需要把握建言方式,有技巧、有策略地向上级进行建言。以往文献对如何建言这一极为重要的问题鲜有研究。本文通过分析50位企事业工作人员的访谈资料发现:建言者在向上级提出建议时,会综合衡量自身(voicer)、对方(object)、内容(issue)、沟通方式(communication)和环境(environment)五个维度,进而采取适当的策略。由此,本文提出员工建言策略的VOICE模型,提炼出建言者定位、建言对象定位、内容架构、沟通优化和环境适配五种策略。这五种建言策略会通过化解情感阻力和增进建言价值认知两个路径影响建言采纳。VOICE模型对解答员工“如何成功建言”这一难题具有参考价值,对组织管理者构建积极、开放的建言氛围亦有启示意义。

The 2018 M_S 5.9 Mojiang Earthquake:Source model and intensity based on near-field seismic recordings

On September 8, 2018, an M_S 5.9 earthquake struck Mojiang, a county in Yunnan Province, China. We collect near-field seismic recordings(epicentral distances less than 200 km) to relocate the mainshock and the aftershocks within the first 60 hours to determine the focal mechanism solutions of the mainshock and some of the aftershocks and to invert for the finite-fault model of the mainshock.The focal mechanism solution of the mainshock and the relocation results of the aftershocks constrain the mainshock on a nearly vertical fault plane striking northeast and dipping to the southeast. The inversion of the finite-fault model reveals only a single slip asperity on the fault plane. The major slip is distributed above the initiation point, ～14 km wide along the down-dip direction and ～14 km long along the strike direction, with a maximal slip of ～22 cm at a depth of ～6 km. The focal mechanism solutions of the aftershocks show that most of the aftershocks are of the strike-slip type, a number of them are of the normal-slip type, and only a few of them are of the thrust-slip type.On average, strike-slip is dominant on the fault plane of the mainshock, as the focal mechanism solution of the mainshock suggests, but when examined in detail, slight thrust-slip appears on the southwest of the fault plane while an obvious part of normal-slip appears on the northeast, which is consistent with what the focal mechanism solutions of the aftershocks display. The multiple types of aftershock focal mechanism solutions and the slip details of the mainshock both suggest a complex tectonic setting, stress setting, or both. The intensity contours predicted exhibit a longer axis trending from northeast to southwest and a maximal intensity of Ⅷ around the epicenter and in the northwest.

A search-free near-field source localization method with exact signal model

Most of the near-field source localization methods are developed with the approximated signal model,because the phases of the received near-field signal are highly non-linear.Nevertheless,the approximated signal model based methods suffer from model mismatch and performance degradation while the exact signal model based estimation methods usually involve parameter searching or multiple decomposition procedures.In this paper,a search-free near-field source localization method is proposed with the exact signal model.Firstly,the approximative estimates of the direction of arrival(DOA)and range are obtained by using the approximated signal model based method through parameter separation and polynomial rooting operations.Then,the approximative estimates are corrected with the exact signal model according to the exact expressions of phase difference in near-field observations.The proposed method avoids spectral searching and parameter pairing and has enhanced estimation performance.Numerical simulations are provided to demonstrate the effectiveness of the proposed method.

Segregation of voiced and unvoiced components from residual of speech signal

In conventional source-filter models, voiced and unvoiced components were considered independently. However, in practice it was difficult to separate the source into two parts. An actual source consists of a mixture of two sources and the ratio varies according to the content or the intention of speaker. It had been investigated to separate the voiced and unvoiced components for different source models. Source signals were modeled based on the residual signal measured from inverse filtering. Three different source models were assumed. The parameters of each model were optimized for the original speech signal using a genetic algorithm. The resulting parameters were compared in terms of the mel-cepstral distance to the original signal, the spectrogram and the spectral envelope from the synthesized signal. The optimization method achieves an improvement of 15% for the Klatt model, but there is little improvement in the modified residual case.

Hybrid Slip Model for Near-Field Ground Motion Estimation Based on Uncertainty of Source Parameters

The hybrid slip model used to generate a finite fault model for near-field ground motion estimation and seismic hazard assessment was improved to express the uncertainty of the source form of a future earthquake.In this process, source parameters were treated as normal random variables, and the Fortran code of hybrid slip model was modified by adding a random number generator so that the code could generate many finite fault models with different dimensions and slip distributions for a given magnitude.Furth...

AN IMPROVED ALGORITHM OF GMM VOICE CONVERSION SYSTEM BASED ON CHANGING THE TIME-SCALE

This paper improves and presents an advanced method of the voice conversion system based on Gaussian Mixture Models(GMM) models by changing the time-scale of speech.The Speech Transformation and Representation using Adaptive Interpolation of weiGHTed spectrum(STRAIGHT) model is adopted to extract the spectrum features,and the GMM models are trained to generate the conversion function.The spectrum features of a source speech will be converted by the conversion function.The time-scale of speech is changed by extracting the converted features and adding to the spectrum.The conversion voice was evaluated by subjective and objective measurements.The results confirm that the transformed speech not only approximates the characteristics of the target speaker,but also more natural and more intelligible.

Robust Speech Endpoint Detection in Airplane Cockpit Voice Background

A method of robust speech endpoint detection in airplane cockpit voice background is presented. Based on the analysis of background noise character, a complex Laplacian distribution model directly aiming at noisy speech is established. Then the likelihood ratio test based on binary hypothesis test is carried out. The decision criterion of conventional maximum a posterior incorporating the inter-frame correlation leads to two separate thresholds. Speech endpoint detection decision is finally made depend on the previous frame and the observed spectrum, and the speech endpoint is searched based on the decision. Compared with the typical algorithms, the proposed method operates robust in the airplane cockpit voice background.

整合类内差异与类间关联的隐喻感情预测

隐喻感情预测有助于改进社交媒体内容的用户体验,同时在心理健康监测和虚拟心理治疗方面也具有潜在价值。此外,它还可以更精确地识别目标受众的感情需求,优化广告策略,提升商业效益。为了进一步提升情绪识别与情感检测的有效性,提出了一种整合类内差异与类间关联的多模式隐喻感情识别架构。首先引出三种单模式模型,包括视觉语义模型,文本语义模型和音频语义模型,从文本、视觉和音频三种不同的数据源中分别提取每个模式的个性化差异特征;随后引出一种深度层次多模式模型,通过中间层融合的方式对多模式之间的关联性进行学习,更好地利用双模式与三模式之间提供的互补信息;最后,基于决策层融合的方式将上述四种模型进行融合,在一种端到端的框架中,实现多模式隐喻感情的预测。在开源的数据集中进行的大量消融实验与对比实验证明了方法的有效性。

基于多重互信息约束的高表现力语音转换

随着语音转换在人机交互领域的广泛应用,对于获取高表现力语音的需求日益显著.当前语音转换主要通过解耦声学特征实现,侧重对内容和音色特征的解耦,很少考虑语音中混合的情感特性,导致转换音频情感表现力不足.为解决上述问题,本文提出一种基于多重互信息约束的高表现力语音转换模型(MMIC-EVC).在对内容和音色特征进行解耦的基础上,引入表现力模块分别对话语级韵律和节奏特征进行建模,以实现情感特性的传递;随后通过最小化各特征之间的多重互信息变分对数上界,约束各编码器专注于解耦对应的声学嵌入.在CSTR-VCTK和ESD语音数据集上的实验表明,本模型的转换音频语音自然度评分(MOS)达到3.78,梅尔倒谱失真为5.39 dB,最佳最差占比测试结果大幅领先于基线模型,MMIC-EVC能够有效解耦韵律和节奏特征,并实现高表现力语音转换,为人机交互带来更加出色和自然的用户体验.

AI语音助手用户虚拟在线体验影响因素

在以ChatGPT为代表的ICT技术快速发展和社会变革环境下,以AI语音助手为例的智能应用对用户体验和行为的影响成为重要研究问题。文章基于认知情感意图(Cognition-Affect-Co⁃nation,CAC)框架,从AI语音助手技术特征出发分析感知拟人性、感知关系性、感知娱乐性对于用户认知评价的感知服务质量和感知使用成本,以及在认知评价作用下的情感体验对用户采纳意愿的影响。利用结构方法和fsQCA混合研究方法,通过对328份受访者的问卷调查进行实证检验。研究发现:AI语音助手的感知拟人性增加用户的感知使用成本,感知关系性和感知娱乐性有利于感知服务质量;情感体验在用户认知评价与用户行为意图结果之间起中介作用;根据fsQCA数据结果总结并诠释5种AI语音助手用户在线虚拟体验组态路径,验证并补充结构方程研究发现。文章为AI语音助手用户虚拟在线体验影响因素研究提供了新的理论研究视角,深化了客户旅程理论在用户虚拟在线体验中理论应用,为各界利用AI语音助手增强用户体验提供参考。

组织文化认同对铁路工程企业员工建言行为的影响研究

为激发铁路工程企业员工建言行为的发生,研究以组织文化认同为切入点,在文献理论分析基础上,基于社会认同理论和自我决定理论,引入情感投入和感知组织支持作为组织文化认同和建言行为间关系的中介变量和调节变量,探究组织文化认同对铁路工程企业员工建言行为影响机理。使用组织文化认同和建言行为等成熟量表,以中国铁道建筑总公司、中国铁路工程总公司,中国交通建设集团有限公司企业员工为研究对象进行问卷调查;运用Mplus8.0和SPSS27.0等分析软件对收集数据进行验证性因子分析、相关性分析;在验证性因子分析和相关性分析的基础上,将建立的结构方程模型通过运用Mplus8.0进行路径假设检验;运用fsQCA3.0进行前因变量对建言行为模糊定性比较分析。结构方程模型检验结果表明,组织文化认同显著正向影响情感投入;情感投入正向影响建言行为;即情感投入在组织文化认同和建言行为影响关系中起中介作用;组织文化认同显著正向影响建言行为;感知组织支持显著正向调节情感投入对建言行为的影响。fsQCA分析结果表明,有5种组态可以发生建言行为,其中组织文化认同3个维度为核心存在条件,与结构方程模型检验结果相符合。以上结果丰富了建言行为边界机制和作用条件,基于“想做”和“能做”二元视角,揭示了组织工作氛围中,组织文化认同对员工产生“想建言”的心理动机产生重要影响,感知组织支持(“能建言”)正向促进了员工将建言的动机转化为实际的建言行为,为铁路工程企业管理者在实践中提供建议和启示。