...报道为峰值信噪比（PSNR）和结构相似性指数测量（SSIM）。实验结果...
Computed tomography (CT) is one of the most widely used radiography exams worldwide for different diagnostic applications. However, CT scans involve ioniz- ing radiational exposure, which raises health concerns. Counter- intuitively, low- ering the adequate CT dose level introduces noise and reduces the image quality, which may impact clinical (2022-04-26, Python, 5854KB, 下载0次)
滴答词典点此翻译查询源码是一套基于安卓的单词、语句翻译应用项目，提供词典翻译、音标(拼音)、解释、相关词条、例句、读音(英文Only)、候选词等功能。增加的语音识别功能。数据查询使用第三方翻译网站的接口。不过因为接口变动导致翻译功能失效了，可以自己根据爱词霸或者海词的接口重新解析XML绑定数据。词海接口：http://dict.cn/hello。
Translate the tick dictionary query source code is a set of Android s words, sentence translation application project based on, providing dictionary translation, phonetic alphabet (Pinyin), interpretation, related entries, sentences and pronunciation (English only), candidate words such as function. Increased speech recognition. Data query using the interface of the third party translation website. But because the interface changes result in the failure of the translation function, can according to their love PowerWord or sea word interface to parse the XML data binding. Word sea interface: http://dict.cn/hello. (2016-07-11, Java, 1254KB, 下载3次)
基于匹配滤波器的语音识别。本程序接触匹配滤波对加噪语音进行匹配输出，使得信号的信噪比得到提升，送入后端DTW的语音识别系统，实验证明，本程序对语音提升有很大的帮助
A speech recoginiton base on the matched filter. this code will use matched filter and get speech SNR output enhanced. then we will extrace MFCC, send the feature to the DTW recongition system, the test result verify that this method is good for speech recoginition SNR enhanced. (2015-01-10, matlab, 4547KB, 下载16次)
我们设计并实现了一种用于实时检测语音基频（pitch）、提取音乐旋律（melody）的软件。
语音或者音乐的基频是一种很重要的特征，可以用在很多研究与应用中。比如：用于普通话语音声调的识别和确认；语音流利度的分析；音乐旋律的提取用于哼唱检索等。读者可以参考我们之前的博文：http://blog.sciencenet.cn/home.php?mod=space&uid=907554&do=blog&id=723808
We design and implement software for real-time detection of voice fundamental frequency (pitch), extract music melody (melody) of.
The fundamental frequency of the voice or music is a very important feature, can be used in many studies and applications. For example: for Mandarin speech recognition and confirmation tone analysis of speech fluency extract melodies for humming retrieval. Readers can refer to our previous blog: http: //blog.sciencenet.cn/home.php mod = space & uid = 907554 & do = blog & id = 723808? (2014-08-05, Visual C++, 1120KB, 下载18次)
核磁共振（NMR）技术探测地下水是目前唯一的直接找水的地球物理方法。与传统的地球物理方法相比具有高分辨力，高效率，信息量丰富和解唯一性等优点。利用核磁共振地下水探测系统可以高效率地进行区域水文地质调查，确定找水远景区，圈定地下水在三维空间内的分布，进而可靠地选定水井位置等。但是由于水中氢核产生的核磁共振信号的幅度小（纳伏级），要求探测系统的灵敏度高，将引入大量的自然和人为噪声，导致采集信号的信噪比低，解释结果不清楚。
本文用LMS自适应算法提高MRS信号的信噪比，用Hilbert变换提取MRS信号的包络信号，用线性拟合提取水文地质参数并通过仿真实验结果，验证了本文提出的数据处理方法的有效性和准确性。
Nuclear magnetic resonance (NMR) technique to the detection of direct water division is the only geophysical method to find water. Compared with conventional high-resolution geophysical methods, efficient, informative unique advantages of reconciliation. Groundwater exploration using nuclear magnetic resonance system can efficiently carry out the regional hydrogeological investigation to determine the prospective areas to find water, delineation of ground water distribution in three-dimensional space, and then reliably the location of the selected wells. However, because the water proton NMR signal amplitude generated by small (Na V level), requiring high sensitivity detection systems, will introduce a large number of natural and man-made noise, resulting in low signal to noise ratio acquisition, interpretation of results is not clear.
In this paper, LMS adaptive algorithm improves the MRS signal to noise ratio, with the Hilbert transform to extract the signal envelope MRS signal ext (2013-05-15, WORD, 932KB, 下载14次)
这是一个中文普通话语音识别的最简单DEMO，LilyMandarinSpeechRecognition V1.0。
首先给出这款软件的下载链接：
http://vdisk.weibo.com/s/AUDeE/1368171209
DEMO详细描述参看http://blog.sina.com.cn/s/blog_6a4085570101bwns.html。语音不仅是人类之间进行信息交流最自然,最有效,最方便的工具,而且也是人与机器之间进行通信的重要工具。语音识别(Automatic Speech Recognition,ASR)技术能够让机器听懂人的声音，并做出正确的反应。本DEMO完全通过HTK工具（http://htk.eng.cam.ac.uk/）搭建。感谢HTK工作组对语音识别研究领域做出的伟大贡献。
This is a Mandarin Chinese speech recognition easiest Demo LilyMandarinSpeechRecognition V1.0. Firstly, this software download link: http://vdisk.weibo.com/s/AUDeE/1368171209 DEMO detailed description see http://blog.sina.com.cn/s/blog_6a4085570101bwns.html. Voice is not only a human between the exchange of information is the most natural, the most effective, most convenient tool, but also an important tool of communication between man and machine. Speech recognition (Automatic Speech Recognition, ASR) technology enables the machine to understand the human voice, and make the right response. The DEMO completely through the the HTK Tools (http://htk.eng.cam.ac.uk/) structures. Thanks for the great contribution made by the the HTK working group on the field of speech recognition research. (2013-05-13, Visual C++, 7291KB, 下载9次)
功能 1 客户在拨打企业电话客服时，系统会自动接收、识别并记录来电的主叫号码并显示相应的客户信息。
功能 2 客户电话打入时系统自动播放已设定好的欢迎词。
用户在拨打企业电话客服后，系统会进行自动语音引导，客户通过按键选择，从而可以听到预先设置好的语音咨讯，如企业简介、信息、最新产品等……。所有咨讯信息通过按键即可获得。
功能 3 外线电话转接到座席电话功能，使来电者可以根据系统提示，转入人工服务，与座席人员直接进行交流。
座席电话转接到座席电话功能
Function 1 customers call the business telephone customer, the system will automatically receive, identify and record the caller number to display the customer information. The AutoPlay function 2 customer call comes in when the system has set a good welcome speech. users call the business telephone customer, automatic voice guide customers through the key choices, so that you can hear the pre-set voice Oration, such as corporate profiles, information, latest products ... All Oration information button. The function outside calls transferred to the agent phone features, so that the caller can be prompted into the labor service, and agents who directly exchange. The seats forward calls to the agent phone features (2012-11-23, Visual C++, 764KB, 下载2次)
采用广义旁瓣抵消（GSC）自适应波束形成方法实现时域和频域滤波，采用LMS自适应算法，最终实现语音增强。（文件中包含纯净语音及不同信噪比的带噪语音）
Generalized sidelobe canceller (GSC) adaptive beamforming method to achieve time-domain filtering using the LMS adaptive algorithm, and ultimately the speech enhancement (2012-09-07, matlab, 201KB, 下载605次)
ITU P.563 语音质量评价，能只送入失真信信号，而不需要参考信号，便能评价失真信号的MOS分
ITU P.563 voice quality assessment can only be sent to the distortion of the letter signals without the need for a reference signal will be able to evaluate distortion signal MOS score (2012-07-11, Visual C++, 1024KB, 下载61次)
论文首先介绍了传统的语音特征参数MFCC，它是基于人耳听觉
特性设计的一种特征参数，在静音环境下能得到较高的识别率，但在
信噪比较低时识别率急剧下降，不利于实用化。本文通过对MFCC算
法的分析和研究，发现其中的FFT和DCT在整个时频空间使用固定的
。分析窗，这不符合语音信号特性，而小波变换具有多分辨率特性，更
符合人耳的听觉特性。因此，本文将小波变换和MFCC算法相结合，
提出了三种新的语音识别特
Speech recognition has wide use in the field of communication and so
on·Speech feature parameter extraction is an important part of the speech
recognition system．The performance of the feature parameter haftuences
the system’S performance directly．And the environmental noise is a kev
factor of restricting the performance of the feature parameter．This paper，s
research object is extracting the speech feature parameter under the noise
environment·Then it analyzes the auditory model of human，deeply
analyzes and researches the traditional speech feature parameter MFCC，
and proposes three improved MFCC parameters based on theⅥ，avelet
transformation and the auditory characteristic ofhuman．Besides，the paper
also gives all improved method about the feature parameter ZCPA． (2011-08-13, matlab, 1955KB, 下载21次)
为提高语音端点检测系统在低信噪(0 dB 以下) 下
检测的准确率, 提出了一种基于谱熵的端点检测算法。将每
帧信号分为16 个子带, 选取频谱分布在250～ 3. 5 kHz 并且
能量不超过该帧总能量90 的子带, 计算经过语音增强后的
子带能量以及各子带信噪比, 根据各子带信噪比的不同调整
其在整个谱熵计算过程中的权重, 然后平滑谱熵, 以最终的
谱熵作为端点检测的依据
To improve endpoint detection system in the low signal to noise (0 dB or less) under the detection accuracy, a spectrum entropy-based endpoint detection algorithm. Each frame signal is divided into 16 sub-bands, select the frequency distribution in the 250 ~ 3. 5 kHz and the energy of not more than 90 of the total energy of the frame of the sub-band, calculated through the following sub-band speech enhancement as well as the sub-band signal to noise ratio of energy , according to the different sub-band signal to noise ratio to adjust its calculation of the spectral entropy of the process of weight, and smooth spectral entropy, spectral entropy to the final endpoint detection as the basis for (2011-05-27, Visual C++, 255KB, 下载30次)
为提高语音端点检测(VAD)在较低信噪比(10 dB)下的准确率,提出一种基于短时分形维数的改进算法。结合语音信号的特点,对2种常用的语音信号分形维数计算方法进行了比较和选择,同时采用动态跟随门限值实现语音端点的自适应检测。试验结果表明:对于信噪比6~10 dB的带噪语音,此方法可以实现整段语音的检测,而且具有一定的噪声鲁棒性,系统运行期间能够自适应调整门限值以适应环境噪声的变化,提高了VAD算法的准确率。这个是源码matlab。
In order to improve voice activity detection (VAD) in low SNR (10 dB) accuracy under proposed based on short-time fractal dimension of the improved algorithm. Combined with the characteristics of the speech signal, to 2 commonly used fractal dimension of speech signals are compared and calculated choice to follow the same dynamic endpoint threshold adaptive detection of voice. The results showed that: 6 ~ 10 dB for the signal to noise ratio of noisy speech, this method can detect the entire speech, but has some noise robustness, the system can be adaptively adjusted during operation to adapt to environmental noise threshold of changes to improve the accuracy of VAD algorithms. This is the source matlab. (2011-04-29, Visual C++, 78KB, 下载70次)
The famous speech signal processing tool
The famous speech signal processing tool (2010-12-03, C/C++, 2979KB, 下载3次)
介绍了改进谱减法的原理及算法。提出在信噪比(sNR)较低的情况下，根据语音短时能量和过零率，判断在无声 或有声期间是否偶然的噪声过高，从而设定合适的参数降低噪声。
Introduced an improved spectral subtraction principles and algorithms. Proposed signal to noise ratio (sNR) lower case, according to voice short-term energy and zero-crossing rate, to judge whether silent or sound during the occasional noise is too high, and thus set the appropriate parameters to reduce noise. (2010-02-12, PDF, 235KB, 下载33次)
利用I.Cohen提出的短时先验信噪比（priori SNR）估算的语音增强法，附有相应程序和文献。
I. Cohen raised by the use of short-term a priori signal to noise ratio (priori SNR) estimation of the speech enhancement method, with the corresponding procedures and documentation. (2008-11-23, matlab, 123KB, 下载133次)
为均衡带限信号所引起失真的横向或格型自适应均衡器(其中横向FIR系统长M=11), 系统输入是取值为±1的随机序列 ，其均值为零；参考信号 ；信道具有脉冲响应： 式中w用来控制信道的幅度失真(w = 2～4，例如，取w = 2.9,3.1,3.3,3.5等),而且信道受到均值为零、方差为 (例如，取 ，相当于信噪比为30dB)的高斯白噪声 的干扰。试比较基于下列五种算法自适应均衡器在不同信道失真、不同噪声干扰下的收敛情况(对应于每一种情况，在同一坐标下画出其学习曲线)： 横向/格-梯型结构LMS算法[1-4] 横向/格-梯型结构RLS算法[1-4] (2007-09-12, matlab, 1KB, 下载108次)
http://www.pudn.com/Download/item/id/332022.html摘要：基于语音状态模型的语音增强算法是当前语音信号处理的研究热点。把通常的LPC语音模型修正后，将得到两个语音模型：时变AR 模型、时变双AR模型。但是利用这些模型增强语音时，都没有考虑到语音的清音、浊音区别。为此本文引入了语音清浊音状态空间模型，这种模型在描述语音方面比时变AR模型、时变双AR模型要强，而且物理含义明显 同时在用含噪语音信号预测纯净语音信号时，引入遗忘因子和粒子滤波算法以降低计算复杂性，减小运算量。实验证明，增强后的语音信号信噪比有一定提高．且优于传统的LPC模型. (2007-09-12, PDF, 170KB, 下载35次)
http://www.pudn.com/Download/item/id/331797.html
谱减法是一种比较简单的语音降噪的方法，在语音信号具有较高的信噪比的情况下具有良好的降噪功能。
Spectral subtraction is a relatively simple method of noise reduction voice in the voice signal with higher signal to noise ratio case has a good noise reduction function. (2007-07-11, PDF, 216KB, 下载196次)
语音端点检测是语音识别中至关重要的技术。无论军用还是民用，语音端点检测都有着广泛的应用。在低信噪比的环境中进行精确的端点检测比较困难，尤其是在无声段或者发音前后
voice activity detection is critical speech recognition technologies. Whether military or civilian, voice endpoint detection have broad application. Low signal-to-noise ratio in the environment for accurate endpoint detection more difficult, especially in or pronunciation of the silent before and after (2007-05-18, matlab, 519KB, 下载153次)
当你的电话语音卡收不到主叫号码的时候，可以用这个程序来测试
步骤
1）选择好通道号
2）选择格式
3）选择开始
4）拨打电话
当程序接收不到主叫的时候，你可以把现场的主叫声音录下来进行分析
步骤
1）选好通道
2）选择录音
3）拨打电话
4）三声回铃声以后，选择停止录音，生成record.pcm主叫声音文件
5）用语音编辑工具打开record.pcm
6）对照网页http://www.dj.com.cn/support/cid/techid.asp的图形来进行分析
When your telephone can not receive Caller ID card, they can use this procedure to test the step 1) the choice of a good channel No. 2) Select the format 3) Select Start 4) When the procedure call when the caller can not receive, You can put the caller at the scene recorded sound analysis steps 1) select Channel 2) Select recording 3) call 4) Back to ring three times, the choice to stop recording, sound files generated record.pcm caller 5) voice editing tool to open record.pcm6) control of the graphics page to http://www.dj.com.cn/support/cid/techid.asp analysis (2006-05-15, Visual C++, 476KB, 下载26次)