资源简介
在matlab下使用一些常见的处理音频的函数,例如enframe等,都需要调用voicebox包,默认在matlab下是不安装的,需要手动加载。加载方法就是在命令行里输入addpath(genpath('E:/voicebox')) 其中路径可以自己改
代码片段和文件信息
function [levaffsovad]=activlev(spfsmode)
%ACTIVLEV Measure active speech level as in ITU-T P.56 [LEVAFFSO]=(spFSMODE)
%
%Usage: (1) lev=activlev(sfs); % speech level in units of power
% (2) db=activlev(sfs‘d‘); % speech level in dB
% (3) s=activlev(sfs‘n‘); % normalize active level to 0 dB
%
%Inputs: sp is the speech signal (with better than 20dB SNR)
% FS is the sample frequency in Hz (see also FSO below)
% MODE is a combination of the following:
% 0 - omit high pass filter completely (i.e. include DC)
% 3 - high pass filter at 30 Hz instead of 200 Hz (but allows mains hum to pass)
% 4 - high pass filter at 40 Hz instead of 200 Hz (but allows mains hum to pass)
% 1 - use cheybyshev 1 filter
% 2 - use chebyshev 2 filter (default)
% e - use elliptic filter
% h - omit low pass filter at 5.5 12 or 18 kHz
% w - use wideband filter frequencies: 70 Hz to 12 kHz
% W - use ultra wideband filter frequencies: 30 Hz to 18 kHz
% d - give outputs in dB rather than power
% n - output a normalized speech signal as the first argument
% N - output a normalized filtered speech signal as the first argument
% l - give both active and long-term power levels
% a - include A-weighting filter
% i - include ITU-R-BS.468/ITU-T-J.16 weighting filter
% z - do NOT zero-pad the signal by 0.35 s
%
%Outputs:
% If the “n“ option is specified a speech signal normalized to 0dB will be given as
% the first output followed by the other outputs.
% LEV gives the speech level in units of power (or dB if mode=‘d‘)
% if mode=‘l‘ is specified LEV is a row vector with the “long term
% level“ as its second element (this is just the mean power)
% AF is the activity factor (or duty cycle) in the range 0 to 1
% FSO is a column vector of intermediate information that allows
% you to process a speech signal in chunks. Thus:
% fso=fs;
% for i=1:inc:nsamp
% [levaffso]=activlev(sp(i:min(i+inc-1nsamp))fso[‘z‘ mode]);
% end
% lev=activlev([]fso)
% is equivalent to:
% lev=activlev(sp(1:nsamp)fsmode)
% but is much slower. The two methods will not give identical results
% because they will use slightly different thresholds. Note you need
% the ‘z‘ option for all calls except the last.
% VAD is a boolean vector the same length as sp that acts as an approximate voice activity detector
%For completeness we list here the contents of the FSO structure:
%
% ffs : sample frequency
% fmd : mod
属性 大小 日期 时间 名称
----------- --------- ---------- ----- ----
文件 16679 2017-02-07 21:25 voicebox\activlev.m
文件 8389 2014-08-06 02:25 voicebox\activlevg.m
文件 1793 2014-01-16 15:43 voicebox\atan2sc.m
文件 2112 2014-01-16 15:43 voicebox\axisenlarge.m
文件 4728 2014-04-25 14:51 voicebox\bark2frq.m
文件 1705 2014-04-25 14:51 voicebox\berk2prob.m
文件 3305 2014-01-16 15:43 voicebox\bitsprec.m
文件 2469 2014-01-16 15:43 voicebox\cblabel.m
文件 2013 2014-01-16 15:43 voicebox\ccwarpf.m
文件 2046 2014-01-16 15:43 voicebox\cent2frq.m
文件 2131 2014-01-16 15:43 voicebox\cep2pow.m
文件 2239 2014-01-16 15:43 voicebox\choosenk.m
文件 1603 2014-01-16 15:43 voicebox\choosrnk.m
文件 14643 2017-04-04 22:30 voicebox\Contents.m
文件 3975 2014-01-16 15:43 voicebox\correlogram.m
文件 3819 2014-01-16 15:43 voicebox\distchar.m
文件 3422 2014-01-16 15:43 voicebox\distchpf.m
文件 3688 2014-01-16 15:43 voicebox\disteusq.m
文件 4641 2014-01-16 15:43 voicebox\distisar.m
文件 3736 2014-01-16 15:43 voicebox\distispf.m
文件 4170 2014-01-16 15:43 voicebox\distitar.m
文件 3601 2014-01-16 15:43 voicebox\distitpf.m
文件 1982 2014-01-16 15:43 voicebox\ditherq.m
文件 2756 2016-01-07 02:13 voicebox\dlyapsq.m
文件 3479 2014-01-16 15:43 voicebox\dualdiag.m
文件 26494 2014-01-16 15:43 voicebox\dypsa.m
文件 6637 2017-03-08 17:38 voicebox\enfr
文件 3906 2014-01-16 15:43 voicebox\entropy.m
文件 3430 2015-03-02 00:02 voicebox\erb2frq.m
文件 7306 2014-01-16 15:43 voicebox\estnoiseg.m
............此处省略236个文件信息
评论
共有 条评论