资源列表
yuyin
- 语音信号的时域、频域分析,包括短时能量分析、短时平均过零率、自相关函数、短时平均幅度差函数等。-Time-domain speech signal, frequency domain analysis, including short-term energy analysis, the average short-term zero-crossing rate, autocorrelation function, such as short-time average magnitude diff
SR
- 语音识别，能够识别文字，并且能打开控制电脑中的软件-Speech recognition, to recognize the text, and can open the control computer software
win_endcut
- 基于短时窗能量的自适应语言端点检测程序。具有一定的抗噪声能力。输入WAV文件名，输出端点检测结果，包括数据和画图显示。-Adaptive voice activitie detection program，resistant to noice .Input with the name of wave file,output with the VAD result ,including data and figure display.
lpc10
- LPC-10声码器，包括源程序和中文注释，也简单介绍其原理。-LPC-10 vocoder, including the source and Chinese comments, also briefly describes the principle.
VB_text_reading
- 用VB实现文本朗读功能，有中英文两个代码，很简单的小程序-With VB text reading function, there are two codes in English, very simple little program
yuyin
- 语音识别代码 文字与语音的调节 语音识别可供参考-Regulating speech recognition voice text and voice identification code for reference
Laplace
- 传统的短时谱估计语音增强算法通常假设语音谱分量相互独立，没有考虑语音谱分量间的相关性。针对这 一问题，该文提出一种新的基于多元Laplace分布模型的短时谱估计算法。首先，假设语音的离散余弦变换(DCT) 系数服从多元Laplace分布，以此利用谱分量间的相关性；在此基础上，利用多元随机矢量的高斯尺度混合模型表 示，推导得到语音DCT系数矢量的最小均方误差(MMSE)估计的解析表达式；并进一步推导了基于该分布模型的 语音存在概率，对最小均方误差估计子进行修正。实验结果表明，该算法
Ma-nguon
- Determine source signal by reverse filter with C# project
PNCC2009KimStern
- Paper: PNCC, Audio, Speech, Machine Learning, Speech recognition.
PNCC2012KimStern
- PNCC paper, Kim, Stern, Machine Learnin.
Synthetic Speech Detector Example
- Detects synthetic speech based on interframe difference of log likelihood of claimed speaker s GMM.
iir
- 语音信号的iir滤波器设计,从带有噪音的信号中提取原始声音.目前，MP3播放器一般功率放大器的工作频率范围就是这个范围。但是大部分有用的和可理解的信息的频率在200到3500Hz之间。所以我们可以在这个范围间滤波，达到使声音可理解的要求。现将数字滤波器的设计指标设为通带截止频率fb=600HZ,阻带频率fc=1200HZ,通带波纹Ap=1dB，阻带波纹As=40dB,要求确定H(z)。-design of the iir filter, get the original voice withou