error in autoencoder function

Hello Everyone ,

i am trying to create speaker identification system with autoencoder and Neural Network .First i am trying to create dataset funciton with voice files.

the code is :

function [xTrainVoices,tTrain]= create_voice_data()
xTrainVoices={};
tTrain=[];
file_num=0;

file_extension='.flac'; 
voice_folder='ali\';
voice_folder_content=dir([voice_folder,'*',file_extension]);
file_number = size(voice_folder_content,1);
for k=1:file_number
 file_name=[voice_folder,voice_folder_content(k,1).name];
 Ivoice=audioread(file_name);
 xTrainVoices{k}=Ivoice;
end
file_num=file_num+file_number;
tTrain=zeros(file_number,2);
tTrain(:,1)=1;
% HUseyinin sesinin veri setine eklenmesi
file_extension='.flac'; 
voice_folder='huseyin\'; 
voice_folder_content=dir([voice_folder,'*',file_extension]);
file_number = size(voice_folder_content,1);
for k=1:file_number
 file_name=[voice_folder,voice_folder_content(k,1).name];
 Ivoice=audioread(file_name);
 xTrainVoices{file_num+k}=Ivoice;
end
file_num=file_num+file_number;
Lbanana=zeros(file_number,2);
Lvoice(:,2)=1;
tTrain=[tTrain;Lvoice];
save voice_dataset_auto;

when i try to use this function with autoencoder on thic code:

load ses_veriseti_auto;
hiddenSize1 = 20;
autoenc1 = trainAutoencoder(xTrainVoices,hiddenSize1, ...
 'MaxEpochs',200, ...
 'L2WeightRegularization',0.004, ...
 'SparsityRegularization',4, ...
 'SparsityProportion',0.15, ...
 'ScaleData', false);
feat1 = encode(autoenc1,xTrainVoices);

ı got this error:

Error using Autoencoder>iConvertToDouble (line 922)

Training data for an autoencoder must either be a matrix, or a cell array of equal size images.

Error in Autoencoder.train (line 510)

[X,visualizationDimensions, trainedOnImages] = iConvertToDouble(X);

Error in trainAutoencoder (line 109)

autoenc = Autoencoder.train(X, autonet, paramsStruct.UseGPU);

Error in denemeautoencoder (line 3)

autoenc1 = trainAutoencoder(xTrainVoices,hiddenSize1, ...

i dont know how to fix this...

any help will be appreciated..