▍1. speakercode1
Speaker Identification using MFCC feature and VQ training, very clear MATLAB code
Speaker Identification using MFCC feature and VQ training, very clear MATLAB code
本文主要研究普通计算环境下语音自动识别(ASR)技术的软件实时实现,基于语音识别的基本原理和过程。在VC环境下建立了一套语音识别系统及其演示软件。实验表明,该系统对特定人情况有较高的识别率和一定的实时性(In this paper, under the computing environment on general automatic speech recognition (ASR) technology to achieve real-time software, speech recognition based on the basic principles and processes. VC environment in the establishment of a voice recognition system and its presentation software. The experiments show that the systems are in particular have a high recognition rate and a certain degree of real-time)
code for voice analysis final code for sapeech recognition
基于hmm的识别程序 识别数字0-9,识别率全对(speech recognition based on hmm number 0-9 all right)
台湾国立中央大学开发的EMD-HHT算法,其中,EMD-HHT的创始人为该中心的主任.(EMD-HHT IS SPECIALLY DESIGNED FOR PROCESSING NONSTATIONARY AND NONLINEAR SIGNALS. IT CAN DECOMPSE SIGNALS AND THEN RECONSTRUCT SIGNALS ACCORDING TO SOME CRITERIA. AFTER THE PROCESS, SIGNALS WOULD HAVE HIGH SIGNALTO NOISE RATIO. IT HAS BEEN APPLIED TO VARIOUS TYPES OF SINALS FROM INDUSTRY AND ACHIEVED GREAT SUCCESS. IT EVOLVES FROM THREE SOURCES: FRANCE SCIENTIST G RILLING, MATLAB FILE EXCHANGE CENTRE BY MR TAN AND TAIWAN NATIONAL CENTRAL UNIVERSITY CENTRE.)
EVC 基于CE的TTS源码,需安装语音库(TTS resource code based on wince)
用于汉语语调识别的程序,可以识别我们说话中的四个声调。已通过测试(Chinese tone recognition for the program, has been tested)
性别识别系统的Matlab实现,比较简单容易理解(Gender Recognition System Matlab implementation, relatively simple easy to understand)
一个通过实际应用的“ 人工鱼群算法”,效果很好的。(Through the practical application of the " artificial fish-swarm algorithm," good effects.)
针对信号自适应chirplet分解未知参数多、实现起来比较困难的特点,文献[1]提出了一种新的chirplet分解快速算法。该算法利用计算信号的二次相位函数,得到其能量分布集中于信号的调频率曲线上的结论,此时通过谱峰检测可同时获得chirplet调频率、时间中心和幅度的估计;然后通过解线性调频技术获得其初始频率和宽度的估计,仿真结果验证了本文算法的有效性。(this code includs one method for chirplet analasis,that is used in voice and ultra-wave analasys,etc.)
自相关函数的应用,进行基音检测,在MATLAB下进行仿真(Application of autocorrelation function for pitch detection, carried out in the MATLAB simulation)
C语言fft和ifdct的函数,可以直接使用(C language functions fft and ifdct can be directly used)
基于非线性主分量分析(PAST)的 自适应盲信号分离 (Based on nonlinear principal component analysis (PAST) Adaptive Blind Signal Separation)
使用子带谱熵进行的端点检测,将谱熵分为几个子带,检测效果不错(The use of sub-band entropy of the endpoint detection, the spectral entropy is divided into several sub-band to detect the effect of good)
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 values derived function [bq, ap] = armahat (rx, p, q))
1.单通道语音增强算法; 2.一种能够快速跟踪噪声变化的,基于噪声统计特性的噪声估计算法,结合谱减法进行消除噪声; 3.算法复杂度适中,可以满足实时性(1. Single-channel speech enhancement algorithm 2. A fast-track noise can change, based on the noise statistical properties of the noise estimation algorithm, combined with spectral subtraction to eliminate noise 3. Algorithm moderate complexity to meet the real-time)