解释 simulink 中的 MATLAB Function 块错误
Interpreted MATLAB Function block error in simulink
我创建如下所示的 simulink 块:
在解释的 MATLAB 函数中我使用了这段代码。
function y1 = fcn(signal)
%% variable declaration
Fm = 100;%length of Frames excursion
framelnc = 100;%length of Frames excursion
Fn = 256;%length of Frames sampling frequency 12.500kHz, frame length 020.5ms
fs = 8000;
x=signal;
%% Preprocessing: Noise Removal
IS= 0.25; % initial silence (noise only)
output=denoise(x,fs,IS);
y = output;
%% Extraction of LFCC Features
feat = lfcc(y,fs);
%% Extract VQ CODE
k=16; % number of centroids required
code1 = vqext(feat, k);
test = mean2(code1);
%% Emotions Recognition
load ('train.mat')
Class = knnclassify(test , feature, type);
if (Class == 1)
%% Display identification results
msg=1;
end
if (Class == 2)
%% Display identification results
msg=2;
end
y1 = msg;
我在单独的 matlab 文件中定义了这些 denoise
、lfcc
、vqext
、knnclassify
函数。但我得到如下错误。
我该如何解决这个问题?
从你的图片我们可以看出,你的数据形式上是二维的[1024x1]!所以我认为,解释型 MATLAB 函数不能使用这种数据类型。
所以我建议使用帧转换块。
P.S。顺便问一下,你为什么要使用解释函数?真的有必要吗?
Note This block is slower than the Fcn block because it calls the
MATLAB parser during each integration step. Consider using built-in
blocks (such as the Fcn block or the Math Function block) instead.
Alternatively, you can write the function as a MATLAB S-function or
MEX-file S-function, then access it using the S-Function block.
我创建如下所示的 simulink 块:
在解释的 MATLAB 函数中我使用了这段代码。
function y1 = fcn(signal)
%% variable declaration
Fm = 100;%length of Frames excursion
framelnc = 100;%length of Frames excursion
Fn = 256;%length of Frames sampling frequency 12.500kHz, frame length 020.5ms
fs = 8000;
x=signal;
%% Preprocessing: Noise Removal
IS= 0.25; % initial silence (noise only)
output=denoise(x,fs,IS);
y = output;
%% Extraction of LFCC Features
feat = lfcc(y,fs);
%% Extract VQ CODE
k=16; % number of centroids required
code1 = vqext(feat, k);
test = mean2(code1);
%% Emotions Recognition
load ('train.mat')
Class = knnclassify(test , feature, type);
if (Class == 1)
%% Display identification results
msg=1;
end
if (Class == 2)
%% Display identification results
msg=2;
end
y1 = msg;
我在单独的 matlab 文件中定义了这些 denoise
、lfcc
、vqext
、knnclassify
函数。但我得到如下错误。
我该如何解决这个问题?
从你的图片我们可以看出,你的数据形式上是二维的[1024x1]!所以我认为,解释型 MATLAB 函数不能使用这种数据类型。
所以我建议使用帧转换块。
P.S。顺便问一下,你为什么要使用解释函数?真的有必要吗?
Note This block is slower than the Fcn block because it calls the MATLAB parser during each integration step. Consider using built-in blocks (such as the Fcn block or the Math Function block) instead. Alternatively, you can write the function as a MATLAB S-function or MEX-file S-function, then access it using the S-Function block.