搜索资源列表
weimin
- 读取语音信号(用matlab的wavread指令),把语音信号分帧、加窗,进行清浊分割,提取基 频,这一部分较简单,自己编程序做。参考文献自己到图书馆期刊网上查找。 提取语音信号的lpc参数,可调用lpcfit.m 程序(我提供,见附件),将源、目标语音的浊音 段的lpc系数进行DTW规整,调用pathita2.m 程序(我提供,见附件)。将规整得到的lpc系数 转换成lsp参数,调用lpcar2ls.m 程序(我提供,见附件), 再进行转换映射,调用matlab 的指令ne
GZ7
- 基于相位法的相关差计,即用程序的形式把相位的差计表示出来-based on the phase difference of law, that is, the procedures in the form of phase margin voiced
basicprocessingofspeech
- 语音信号处理的最基本的Maylab处理程序,包括读入语音波形,清音浊音的检测,加窗,过零率,短时能量,基音最大值。最后有test给出演示。-speech signal processing Maylab the most basic procedures, including reading the speech waveform. voiceless voiced detection, adding windows, the zero-rate, short-term energy, the
lsp3
- 语音信号处理的最基本的Maylab处理程序,包括读入语音波形,清音浊音的检测,加窗,过零率,短时能量,基音最大值。最后有test给出演示-speech signal processing Maylab the most basic procedures, including reading the speech waveform. voiceless voiced detection, adding windows, the zero-rate, short-term energy, the m
Speech Enhancement Based On AUnvoiced-Voiced Model
- 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.
基于短时傅里叶变换实现对语音频谱的分析
- 基于短时傅里叶变换实现对语音频谱的分析 (分别对清音和浊音进行分析)。,Based on short-time Fourier Transform of the voice spectrum analysis (respectively, pairs of voiceless and voiced for analysis).
sogram
- 语谱图程序,用于语音信号分析。实现获取基音周期,区分清音、浊音等功能-Spectrum language program for speech signal analysis. Achieve access to the pitch period, the distinction between voiceless, voiced etc.
yy
- 提取短时能量,短时过零率,清音/浊音自相关函数-Extraction of short-term energy, short-term zero crossing rate, voiceless/voiced autocorrelation function
语音信号处理的最基本的Maylab处理程序
- 语音信号处理的最基本的Maylab处理程序,包括读入语音波形,清音浊音的检测,加窗,过零率,短时能量,基音最大值。-The basic speech signal processing Maylab handling procedures, including reading speech signals, voiceless voiced detection, windowed, zero crossing rate, short-term energy, pitch maximum.
AMDF_MatlabAnalysis
- 自相关虽然能够判断清/浊音,并估计浊音的基音周期,但是其运算量很大,用乘法运算所需时间较长。为了避免乘法,采用短时平均幅度差函数(AMDF),验证AMDF与自相关有类似的作用,即平均幅度差函数能够代替自相关函数进行语音分析。-Although autocorrelation can differ from the Voiced/Unvoiced and voiced the pitch period estimate, but the operation is large enough, wit
design_for_test
- Testability is the concern most often voiced by Texas Instruments (TIä ) application specific integrated circuit (ASIC) users. This document is intended to consolidate TI policies into a coherent approach to designing for testability. It is
mfccvaddtw
- 语音信号处理的最基本的Maylab处理程序,包括读入语音波形,清音浊音的检测,加窗,过零率,短时能量,基音最大值。最后有test给出演示-Speech Signal Processing Maylab the most basic treatment procedures, including read into the voice waveform, voiceless voiced detection, add windows, zero-crossing rate, short-term
speech
- 录制音频功能,病可以对语音自动分频以及识别语音的浊音清音-Audio recording function, disease can be automatic voice recognition and voice-band voiced voiceless
UVS
- classification algorithm among Unvoiced/voiced speech and silence
pitch_period_estimation
- This a code to determine the pitch period of voiced speech.-This is a code to determine the pitch period of voiced speech.
accurately_pitch_estimation
- with this mscr ipts we are enable to perform accurately pitch detection.first with short term zerocrossing and short term energy estimation analyses we find voiced regions and then with hps algorithm, find pitch values of utterance.
lpc222
- 这是一个语音识别系统的代码。所用到的方法是线性预测法,基于音节激励和声音信号激励的-This is a rcode about speech identification and the methods used is Linear predictive coding(both Pitch and voiced excited)
scripts
- praat scrips for finding voiced limits due to intensity. It explodes in the end...:)
dereverber
- dereverberator code for matlab using voiced-unvoiced classification of speech signals
qingzhuoyin
- 本设计采用了一种基于三参数组合的方法对语音信号进行了清/浊音判决。在传统两参数(短时能量和过零率)算法基础上增加了自相关函数作为判决参数,减少了清/浊音信号误判率。-This design uses a combination of methods based on three parameters of the voice signals of the Qing/voiced sentence. In the traditional two parameters (short-term ene