基于TC8830AF的智能语音卡结构与设计

本文描述一种基于８０３１单片机和ＴＣ８８３０ＡＦ语音处理器的ＰＣ机智能语音卡，通过软硬件设计解决了ＰＣ－８０３１－ＴＣ８８３０ＡＦ结构中的通讯问题。该语音卡在进行语音的合成、编辑和播放时，仅占用极少的ＰＣ机ＣＰＵ处理时间，能满足计算机系统进行实时语音合成的要求。     

单麦克风和双麦克风语音增强系统的研究

本文对采用减谱法和箱关法的语音增强系统进行了研究，系统的噪声为白噪声和普通机房噪声。减谱法采用单麦克风输入，试验中发现语音的分帧方法对消噪影响很大，通过比较得到了一种具有明显消噪效果的分帧方法，在相关法中采用双麦克风输入，对两个输入信号之间的时延差进行了补偿，在求维纳滤波器系数时，对信号谱进行统计平均，从而使消噪后的语音信号具有较好的自然度。在研究以上两种方法的基础上，我们提出了一种混合消噪系统，     

一种基于SBC技术的16kbit/s高效语音编译码器

本文介绍一种基于SBC技术的16kbit／s高效语音编译码器．在该编码方案中，IIR型正义镜象滤波器（QMF）用于实现语音分带，而基于快速搜索技术的矢量量化（FVQA）用于对除基带（采用ADPCM技术）外的其余高频带语音编码。一片TMS320C25芯片实现了两路16kbit／s语音编译码器，并借此实现了两人双向实时通信。得到良好的语音可懂度、自然度及客观评测结果。     

Influence of body position on breathing and its implications for the evaluation and treatment of speech and voice disorders

This paper examines how breathing differs in the upright and supine body positions. Passive and active forces and associated chest wall motions are described for resting tidal breathing and speech breathing performed in the two positions. Clinical implications are offered regarding evaluation and treatment of breathing behavior in clients with speech and voice disorders.     

LPC语音合成技术及其在微机上的应用

本文对语音合成技术作了概述，介绍了语音发声器官的简化模型，并对其在语音信号的分析过程中给出描述。文中介绍了几种语音合成方法，重点讨论了线性预测方法（ＬＰＣ）及其在语音合成中的应用，并结合ＴＳＰ５２２０给出了应用实例。     

用时域谐波定标的数字语音编码

本文给出了一种以9.6Kb/s数据率实时工作的新的数字语音编码器.该系统由时域谐波定标(TDHS)及子频带编码(SBC)两种不同的语音压缩算法组成.阐述了TDHS的基本原理,简述了几种不同的音调检测方法.本文最后对用DSP硬件实时执行该系统的几种方式作了讨论.     

基于CT技术搭建教务成绩查询热线

以“湖北工业大学教务成绩查询热线”服务系统的设计为实例，讨论了呼叫应答中心服务系统，并与单纯基于互联网的客户服务进行了比较。着重从系统的设计思路、系统软硬件、系统功能模块，及系统运行和管理等方面进行了深入阐述。     

Speech acts, commitment and multi-agent communication

The principle aim of this paper is to reconsider the suitability of Austin and Searle’s Speech Act theory as a basis for agent communication languages. Two distinct computational interpretations of speech acts are considered: the standard “mentalistic” approach associated with the work of Cohen and Levesque which involves attributing beliefs and intentions to artificial agents, and the “social semantics” approach originating (in the context of MAS) with Singh which aims to model commitments that agents undertake as a consequence of communicative actions. Modifications and extensions are proposed to current commitment-based analyses, drawing on recent philosophical studies by Brandom, Habermas and Heath. A case is made for adopting Brandom’s framework of normative pragmatics, modelling dialogue states as deontic scoreboards which keep track of commitments and entitlements that speakers acknowledge and hearers attribute to other interlocutors. The paper concludes by outlining an update semantics and protocol for selected locutions. Rodger Kibble is a Lecturer in the Department of Computing, Goldsmiths College, University of London. He has worked as a researcher at the Information Technology Research Institute, University of Brighton, and the School of Oriental and African Studies, University of London. He received his PhD from the Centre for Cognitive Science in the University of Edinburgh in 1997. He has published conference papers and journal articles in the formal semantics of natural language, natural language generation, anaphora resolution, dialogue modelling, argumentation and multi-agent communication; and coedited Information Sharing: Reference and Presupposition in Language Generation and Interpretation (CSLI, 2002).     

语音信号处理中TLC32044与TMS320C30的接口方法

本文叙述了在语音信号处理中使用ＴＬＣ３２０４４与ＴＭＳ３２０Ｃ３０时的接口方法,给出了经过验证的Ａ／Ｄ程序实例,其他应用领域可由此得到借鉴。     

门控卷积神经网络多通路声学回声消除算法*

多通路声重放系统能够增强听者的现实感与空间感,但在免提通信条件下,其不可避免会受到噪声和回声干扰,严重影响通信质量。针对上述问题,本文提出了一种基于门控卷积循环神经网络的多通路声学回声消除和噪声抑制方法。该方法以传声器接收信号和重放声道的压缩复数谱为网络输入,以近端语音的压缩复数谱为网络的输出目标,直接从传声器拾取信号中恢复近端纯净语音,无需对声重放信号进行去相关处理,解决了传统自适应滤波方法中存在的非唯一解问题,同时保证了多通路声重放质量。仿真和真实声学环境实验均表明本文所提出的方法可显著消除多通路声重放系统的噪声和回声,在语音质量和回声返回衰减增益方面均优于传统算法。