资源列表
hmm_model
- inicializacion, entrenamiento y validacion de modelos hmm para reconocimiento de digitos
mysubspectrum
- 频谱减算法实现分离语音噪声，该程序具有一定的鲁棒性，读取语音数据时，采用了队列缓存的方法等-Spectrum subtraction algorithm to achieve separation of voice noise , the program has a certain robustness , reading the voice data , the cache queue .
wannianli
- 程序设计的万年历设计，采用C++开发环境，能运行-a C program calendar
recognition
- 这是一个语音识别系统，能够识别20个英文单词，匹配度极高-This is a speech recognition system can identify 20 words in English, a high degree of matching
ffmpeg_pcm2aac
- 在linux上通过ffmpeg+faac库把pcm格式音频转换成aac格式音频-On linux through ffmpeg+faac library to convert pcm format audio aac format audio
DTW
- Dynamic Time Warping mATLAB code
Basic_LMS
- 一个用于语音信号的LMS算法的小程序，可以直接运行，但需要有音频文件
gFURLS
- 振动抑制的matlab程序源码 rls算法 matlab程序-Matlab program for Vibration Suppression of rls algorithm matlab source program
t1_sampling
- 这是一个关于信号采样的程序。采样定理要求采样频率大于信号频率2倍以上，否则将不能得到正确的结果。-This is a sample program on the signal. Sampling theorem requires sampling frequency more than 2 times greater than the signal frequency, or they will not get the correct results.
pitch321
- 存放每次移位后的帧的基音周期，加窗，提取一帧数据，每一帧的长度-Storing each frame in the shifted pitch, windowing, extracting a data frame, the length of each frame
lmsandrls
- lms adaptive filter for removing noise from speech
mispecgramm
- 一个能实现对语音信号分帧的STFT和ISTFT的函数