资源列表
mfccc
- MFCC参数很好的反映了人耳的听觉特性,所以在语音识别中我们也用到他,本代码是用matlab仿真了MFCC 参数的提取过程.-MFCC parameters well reflected the auditory characteristics of the human ear, so in speech recognition, we also used him, and this code is used matlab simulation parameters of the MFCC ex
pr_hmm-(1)
- It is a profile to initial the transaction and emission matrix of HMM process.
vad3
- 用双门限检测法做语音分割，程序比较具体，并且能够直接仿真出图，只要把输入语音自己录制一下就可以了，是我自己做的，希望您多多指教-dual-threshold detection so segmentation, procedures more specific, and can map out a direct simulation. As long as the importation of voice recording what they could, yes I do hope you
Downloads
- FSK demodulation telling how to do it after AWGN noise.
noise_vad
- 语音增强 维纳滤波 VAD语音激活检测估计噪声-Wiener filtering speech enhancement estimate the noise VAD Voice Activity Detection
wave read
- 输入一wav格式声音信号 输出对声音信号进行抽取并插值后的频谱图(源程序为8倍重采样)，并可使用play函数进行播放试听，如play(nearest),play(ori) 有ori原始音质、ups插值音质、downs抽取音质三种及以下五种重采样音质 重采样试听种类有nearest,linear,spline,pchip,cubic，插值算法详见interp1函数 可用过取消注释“高斯白噪声”、“高频噪声”对声音信号加入噪声-Wav format input-output
nafise
- neural network to number recognition
pujian
- 基于matlab编程实现带噪语音信号去噪-failed to translate
Voice_activity_detection
- 采用能量门限的语音活动端点检测程序，计算复杂度低．-using energy threshold voice activities endpoint detection procedures, low computational complexity.
vad
- 一个简易的端点检测程序，已经实现，包括基于过零率、短时能量的双门限判决方法。-a simple endpoint detection procedures has been achieved, including on the basis of zero rate Short-term dual energy threshold judgment method.
voice
- 这个 是语音信号处理 里面 短时分析 综合程序
shangFunc
- 基于熵的端点检测算法，比传统的端点检测能更精确的检测出端点-endpoint detection algorithm than the traditional endpoint detection more accurate detection of Endpoint