Question

高手帮帮忙，将下面四段话准确翻译成英文，请不要用翻译软件，谢谢~~~（这是通信专业英语相关翻译）。

1、说话人识别作为生物认证技术的一种,是根据语音波形中反映说话人生理和行为特征的语音参数,自动鉴别说话人身份的一项技术。本文通过分析语音特征参数的特点和说话人识别的基本方法，实现了以MFCC参数，利用动态时间归整算法的与文本相关的说话人识别系统，然后改变了提取的特征参数，以一阶差分MFCC参数为特征，提高了系统的识别率。
2、本文针对视频序列中的运动人脸，提出了一种基于视频分割的帧间差分并结合人脸肤色信息的算法对运动的人脸进行检测。该方法首先利用帧间差分在连续的图像序列中两个或三个相邻帧间采用基于像素的时间差分并且阈值化来识别出图像中物体的运动区域，再利用肤色模型信息检测出运动区域的人脸。经实验表明本文提出的算法对视频序列中的运动人脸有较好的检测效果。
3、语音增强是指从含噪语音中提取尽可能多的纯净的原始语音信号，其目的主要有两个：一是改善语音的清晰度；二是提高语音的可懂度。常见的语音增强算法有维纳滤波法、噪声对消法、减谱法等。其中谱减法具有适应信噪比范围大、方法简单、易于实时处理等优点，但用于语音增强具有明显“音乐噪声”的缺陷。 本文实现了一种基于改进谱减法的语音增强算法。针对传统谱减法的缺陷，采用语音端点检测结合组合递归平滑的方法来估计噪声谱。首先对带噪语音信号进行预处理，然后对其进行FFT变换求频谱，在频域将带噪语音的功率谱减去估计噪声的功率谱得到纯净语音功率谱，开方后就得到语音幅度谱估计，将其相位恢复后再采用IFFT变换恢复增强语音信号，并使用matlab语言编程实现整个算法。仿真结果表明该算法在去除背景噪声的同时，保证了较小的语音失真、提高了语音质量，达到了较好的测听效果。
4、直接序列扩频通信技术由于具有隐蔽性好、抗干扰能力强、抗多径干扰、容易实现码分多址等优点而在无线数据通信、移动通信等领域得到越来越广泛的应用。本文是基于MATLAB的直接序列扩频通信系统仿真。在发送端输入的数字信号信息，先由扩频码发生器生成的扩频码序列调制数字信号以展宽信号的频谱，展宽的信号再调制到射频发送出去。在接收端接收到的信号进行解调，然后由本地产生的与发送端相同的扩频码序列去相关解扩。从结果看出即使在信噪比很大的情况下，仍可以进行正常通信，验证了直接序列扩频通信系统的抗干扰性能。

Answer 1

1, speaker recognition as a kind of biological authentication technology, according to the report in speech waveforms is physiological and behavioral characteristics to the speech parameters, the identity of the speaker automatic identification technology as a. This paper through analyzing the characteristics of speech characteristics parameters of basic and speaker recognition method, has realized the MFCC parameters by using the dynamic time GuiZheng algorithm with text related speaker recognition system, then changed the extraction of characteristic parameters, MFCC parameters in a order difference for characteristics, improve the system recognition rate.
2, this paper video sequence, the sports face is proposed based on video integral frame difference and combined with the face skin color information of face of movement on algorithm testing. This method firstly in the continuous frame difference in image sequences two or three
3, speech enhancement refers to extract from speech with noise of as much as possible the purity of the original speech signal, it aim to have two: one is to improve the definition; speech Two is to improve the speech intelligibility. Common speech enhancement algorithm has wiener filtering method, noise reduction of XiaoFa, such as spectrum method. One has to adapt to spectral subtraction SNR range, the method is simple, easy to real-time processing wait for an advantage, but for speech enhancement has obvious "music noise" defects. This paper based on improved realized a spectral subtraction speech enhancement algorithm. Aimed at the defects of traditional spectral subtraction speech endpoint detection combination, the combination of recursion smooth methods to estimate noise spectrum. First to the speech signal with noise, and the pretreatment on FFT spectrum, ask in frequency conversion
4, direct sequence spread spectrum communication technology because of latent good, strong anti-jamming capability, fight multipath interference, easy to realize the advantages and cdma wireless data communication, mobile communication and other areas to be more widely. This paper is based on the MATLAB simulation of direct sequence spread spectrum communication system. After the sender of digital signal input information, first by spread spectrum code generator generated spread spectrum yards digital signal to the detuning sequence modulation signal spectrum of the signal, the detuning send out again to radiofrequency modulated. In the receiver received signals were demodulated by local produce, then with the sender of the same spread spectrum yards sequence to relevant solution extender. From the results that even in SNR greatly circumstance, still can normal communication, and verifies the direct sequence spread spectrum tong ?

追问

和有道词典翻译出来的一样，这样的文献应该是被动语态才对啊~

Answer 2

这么大段专业文字，专业人员一般没空上百度回答问题。

Answer 3

说句势利些的，您这几段基本该付钱了。

追问

能翻译一段也可以的啊~~要求不高~