语音的

Reconstruction of normal speech from Chinese whispers can be applied in several fields, such as the private speech communication in public, reconstruction of normal speech for the aphonic individuals, the special need for the forensic work and etc.

耳语音转换为正常音，可应用于公共场所的通信、失语者的语音恢复和司法工作的某些特殊需要等方面。

However,until now the existence of cognates and the basic rules of historical comparative way have still deserved doubts.

历史比较法的基本原则，即语音相似、语义相近，在确定同源词的情况下才有可能贯穿于寻找语音对应规律和构拟原始形式。

Division I is part of the United Kingdom Elda Group, affiliated to the Chinese mainland, Futian District, Shenzhen, Guangdong Binhe a kerbside domestic subsidiaries, Shenzhen City×Technology Co., Ltd., is set research and development, production, sales, in wireless voice over professionals in the field of manufacturing businesses, the company mainly engaged in wireless voice transmission, chips, components, modules and finished products, OEM, ODM.

我司隶属于英国艾尔达集团旗下，所属中国大陆在於广东深圳市福田区滨河路旁的一家国内子公司，深圳市威雷斯科技有限公司，乃是集研发，生产，销售于一体，在无线语音传输领域里的专业制造商家，公司主要经营无线语音传输，晶片，元件，模组，成品，OEM，ODM。

Real-time speech/music classification is very important in combined speech and audio coding.

语音／音乐的实时分类，在语音／音乐通用编码器中是尤其重要的。

The built-in, multi-waveform modem and 600/2400 bps vocoders provide high data throughput and secure digital voice over the most challenging HF channels.

内置多波形调制解调器和600/2400bps语音编码器提供高数据传输量，通过最具挑战的高频通道进行安全的数字语音。

In DTW matching system, it is very difficulty to determine the cluster center of the speech reference patterns obtained by multi-training.

对于采用DTW动态规划模式匹配方法的语音识别系统，从多次训练所得的语音参考模式中求得其聚类中心一直是一个难题。

Perception of two similar vowels is not confined by the bilinguals' assimilation of production.

相似音的语音感知并不会受到语音生成类化的局限。

This could lead to performance degradations. How to provide a reliable VoIP architecture with the guarantee of quality of voice data is an urgent research topic.

因此，如何建立一个稳定及具有语音封包服务品质保证的网路语音电话架构，是一个值得研究的课题。

A new method of speech emotion recognition via voting combination of multiple classifiers is proposed for improving speech emotion classification rate.

为了提高语音情感的正确识别率，提出一种基于多分类器投票组合的语音情感识别新方法。

We also annotate a speech database of all Taiwanese monosyllables with their pitch marks that is useful for Text-To-Speech synthesis application.

我们同时建立一个可以被用来做语音合成的台语单音节带著正确基周标位的语音资料库。

She gently rebuff ed him, but agreed that they could be friends

她婉言拒绝了，但同意作为朋友相处。

If in the penal farm, you were sure to be criticized.

要是在劳改农场，你等着挨绳子吧！