In this project, we used voice command to control the movement of the car by processing the voice commands in MATLAB and maintaining MATLAB-ARDUINO communication via Bluetooth module.
Below the matlab code for prerecording MFCCs for audio commands is shown.
clc; clear all;
Fs = 44100; nBits = 8; nChannels = 1;
Tw = 20;
# analysis frame duration (ms)
Ts = 10;
# analysis frame shift (ms)
alpha = 0.97;
# preemphasis coefficient
R = [ 300 3700 ]; % frequency range to consider
M = 20;
# number of filterbank channels
C = 13;
# number of cepstral coefficients
L = 22;
# cepstral sine lifter parameter
# hamming window
hamming = @(N)(0.54-0.46*cos(2*pi*[0:N-1].'/(N-1)));
A = [];
for f = 1 : 4
n = input( 'For how many sec you want to record?' );
recObj = audiorecorder(Fs,nBits,nChannels);
disp( 'Start speaking.' );
recordblocking( recObj,n );
disp( 'End of Recording.' );
play(recObj);
y = getaudiodata(recObj);
[ b, FBEs, frames ] = mfcc( y, Fs, Tw, Ts, alpha, hamming, R, M, C, L );
A(:,:,f) = b;
endThe user given command at test time was aligned and the closest match was found using Euclidean distance. The code snippet for this alignment and finding closest match is given below.
# Connection of Bluetooth module
a = Bluetooth('HC-05',1);
fopen(a);
'Bluetooth Successfully Connected'
# record audio for 1 second
recObj = audiorecorder(Fs,nBits,nChannels);
disp( 'Start speaking.' );
recordblocking( recObj,1 );
disp( 'End of Recording.' );
play(recObj);
y = getaudiodata(recObj);
[ b, FBEs, frames ] = mfcc( y, Fs, Tw, Ts, alpha, hamming, R, M, C, L );
# the mfccs for the test command is aligned and matched with the prerecorded audio commands mfccs
error = zeros(4,length(b));
for k = 1:4
for j = 1:length(b)
error(k,j) = dtw(A(:,:,k),circshift(b,j-1,2));end
end
err = min(error,[],2);
result = find(err == min(err))
fprintf(a,result);More details can be found at this link