搜索资源列表
语音识别的前期研究
- 本课题为给语音识别系统提供语音信号端点检测的功能,对语音信号端点检测进行了探讨和研究,主要是侧重语音区间的端点检测...(附软件源码)-Voice Active Detection (VAD) is discussed in this project. Detection between speech section and speechless section and emphasized.(code available).
端点检测程序
- 语音端点检测,是语音压缩时提取特征参数的一个必要的过程,这个程序主要是对这个过程的一个小的程序。希望对学语音的有帮助-Extreme point detecting is a primily process to get the eigenparameter when you compressing the speech.This program is a small one,it will help for the learner.
清浊音检测
- 语音信号的清浊音的提取是研究语音信号的一个必经的过程,如何提取清浊音也是很多初学者面临的问题-Detecting the voice and unvoice is a primily process to study speech signal.How to detect it is also the big problem to the most learner.
Lpcanddtw
- 本程序运用matlab做的语音识别,包含语音检测,滤波-the procedures used Matlab to do voice recognition, including voice detection, filtering, etc.
speechVAD
- 活动语音检测,语音开发中的重要算法,尤其是在voip中广泛使用。该算法简单明了。-voice activity detection, voice development of the algorithm, especially in the widely used voip. The algorithm is simple and clear.
xiaobohezixiangguan
- 本文提出了一种利用二进制小波变换和短时自相关函数相结合的算法来检测语音信号的基音周期。它克服了单纯的短时自相关函数法对含噪语音检测时阈值难以确定的缺陷。-This paper proposes the use of a binary wavelet transform and short-time auto-correlation function of combining voice algorithm to detect letter Pitch No. cycle. It overcome
J005241
- 主要 研 究 用于声音通讯系统中的声回波抵消技术。声回波抵消通常采用声回波抵 消器来实现。具体方法是用自适应滤波器来估计回波信号,并从麦克风信号中减掉该 估计值,从而实现声回波的抵消。回波抵消器中常用的几种自适应滤波算法有LMS, NLMS, RLS等算法。对现有主要算法的性能进行了分析,并对优缺点进行评价和比 较。为了在收敛速度和运算量之间得到很好的折衷,对NLMS算法改进,得到了 PNLMS算法,对传统的快处理块算法改进,得到了精确块算法。综合两个算法,获 得了一种基
speech
- 本文主 要讨沦了语音检测的一些常用方法,并采用自相关方法进行了 模拟实验
红外语音控制
- 三路红外线探头控制ISD2560语音播放。主要原理是正常情况下三对红外线接收管接收到各自配套对应的红外发射管的红外线,属于正常检测状态,一旦有一路红外管信号被隔断,侧启动语音芯片程序,播报警示提醒语音或报警。三对红外管可以根据使用需要,设置搭配报警程序。可用于各种排队系统。-three-way infrared probe control ISD2560 voice broadcast. The main principle is that under normal circumstances
语音信号端点检测的实验研究
- 语音信号端点检测的实验研究论文,希望对大家有点帮助-voice signal endpoint detection of experimental studies, we hope to help a bit
自动语音系统
- 自动语音系统,用于板卡或电信交换机的通话录音与语音检测。次包包括ivr应用系统和二次开发函数接口和实例
matlab语音的端点检测
- 端点检测
duandianjiance
- 自己编的,带有VAD语音检测的谱减法语音增强程序。-Own, and with VAD voice detection spectral subtraction speech enhancement procedures.
endpointdetection
- 语音检测的目的是在一串连续的记录信号中将重要的信息分离出来。在电信应用领域,语音检测是必需的。对于自动语音识别,端点检测在分离重要语音信息时是必需的,这样可以产生语音模式或语音模板。-Voice detection is aimed at a string of consecutive records of information important signal will be separated. At the field of telecommunications applications
nengliang_E
- 这个是基于多带谱熵的能量语音检测,这种处理的效果很好,比传统的谱熵检测效果还明显!-This is based on multi-bands Entropy energy voice activity detection, such an approach worked well, than the traditional spectral entropy test the results were obvious!
duodaipu
- 这个是多带谱语音检测,多带谱不仅具有谱熵的有点,还具有能量分析的有点,所以检测的效果很好!-This is a multi-band spectra of voice activity detection, multi-spectral bands not only has a bit of entropy, also has a bit of energy analysis, so the effect of a very good test!
实现语音的基音检测
- 小波变换,实现语音的基音检测,主要用来检测语音基音周期,小波变换的实现-Wavelet transform, implemented voice detection, voice pitch is mainly used test cycle, the realization of wavelet transform
vad_osf
- 通过计算短时信号熵,与阈值比较,得到判决结果。并采用顺序统计滤波器(最大值滤波器)对所得熵值滤波,提高判决精度。 能有效的分辨出语音信号和噪声信号,且通过后端平滑处理,使语音检测结果更加集中。-VAD based on short-time entropy, precision improved by using order-statistical-filter, VAD result is smoothed to make the result more accurate.
无语音读取简化版本
- matlab实现基于模板的车牌识别.主要模块如下:颜色信息提取、车牌区域定位、识别、提取、检测倾斜度、车牌校正、车牌区域2值化、擦除干扰区域、文字分割、模版匹配、结果输出。(Matlab based template based license plate recognition.The main modules are as follows: color information extraction, license plate location, recognition, extractio
语音基频识别
- 男女生识别、文字端点检测 sgn.m是判断符号的函数 在使用GUI时输入的数字必须为英文状态下(搜狗输入法英文) 端点检测和基频计算是在GUI m文件转到GUI中,基本没什么变化,主要就是使用了全局变量。(Gender recognition and text endpoint detection SGN. M is a function of judging signs When using GUI, the input number must be in English Endpoin