资源列表
efica.rar
- 该程序是FastICA算法的扩展,该程序比原来的FastICA具有更好的分离效果。,The program is an extension of FastICA algorithm, the program than the original FastICA has better separation efficiency.
dtwvc.rar
- 用vc编写的dtw算法,调试通过,用于手写签名识别的,Vc prepared using DTW algorithm, debug is passed, for the handwritten signature recognition
HTK-samples-3.4-alpha.tar.gz
- 语音识别工具HTK最新版,由英国剑桥大学开发,HTK Speech Recognition tools latest version, developed by the University of Cambridge
HMM.rar
- 自己用c语言写的关于语音识别技术中最为流行的匹配算法hmm,相当有用,C language used to write their own on the voice recognition technology in the most popular matching algorithm hmm, very useful
729A.rar
- 用g729实现的编解码,能够用于语音合成等,可以将PCM格式压缩,也可以对压缩后的格式进行解压,Achieved using G729 codec, can be used for speech synthesis, etc., can be compressed PCM format, but also can be compressed format decompression
final.rar基于PC声卡和matlab的实时语音信号采集分析系统
- 一个基于PC声卡和matlab的实时语音信号采集分析系统,带有界面图形,功能较完善,A matlab based on PC sound card and the real-time voice signal acquisition analysis system with graphic interface, function better
LPC-based-voice-spectrum-analy
- 基于语音频谱分析的重要性以及众多端点检测的算法以示其重要性,Based on the importance of voice spectrum analysis and the various endpoint detection algorithm to show its importance
RLS_LMS.rar
- LMS算法、RLS算法 及其二者的比较 有注释说明的 比较详细,LMS algorithm, RLS algorithm and the two more notes of a more detailed
WienerScalart96.rar
- 用Wiener滤波器进行的语音增强算法,效果还可以。 函数输出为WIENERSCALART96(signal,fs,IS),Using Wiener filter speech enhancement algorithm, the effect can be. Function output WIENERSCALART96 (signal, fs, IS)
arma.rar
- ARMA模型建立程序,给出自相关序列,指定ARMA模型的P阶,Q阶参数,得出相应的参数值 function [bq ,ap]=armahat(rx,p,q),ARMA model for the establishment of procedures, to come from related sequences, designated the P-order ARMA model, Q-order parameters, the corresponding parameter value
activlev.rar
- calculates the active level of a speech segment according to ITU-T recommendation P.56.,calculates the active level of a speech segment according to ITU-T recommendation P.56.
voice_recognition.rar
- 使用自相关法提取基频法来判别男女声,并与带通滤波器幅度筛选法及幅度作差提取基频判别法分别作比较。最后的判别正确率接近90%。,The use of autocorrelation-based frequency extraction method to determine male and female voices, and with band-pass filter magnitude and rate of screening for differential extraction of