alu/ece45-project
Archived
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Merge pull request #24 from ltcptgeneral/Nicolas-Schluep-Branch

Browse Source 
Nicolas Schluep Functions Submission
This commit is contained in:
Nicolas Schluep 2021-12-08 15:33:48 -08:00 committed by GitHub
commit e95ff67e8d
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
4 changed files with 173 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

BIN
src/Strong_Bassline.mp3 Normal file
View File

Binary file not shown.

55
src/epic_effect_schluep.m Normal file
View File

@ -0,0 +1,55 @@
function output = epic_effect_schluep(input, Fs, LOW, MED, HIGH)
% epic_effect_schluep: Outputs a more "epic" version of the input sound.
% This is done by amplifying low frequencies and by implementing a chorus
% effect. A chorus effect is created when multiple copies of sound delayed
% by a small, random amount are added to the original signal. Works best on
% songs with a stronger bassline.
% Try this function out with "Strong_Bassline.mp3".
% CONTRIBUTORS:
% Nicolas Schluep: Function Author
% DOCUMENTATION:
% input: The input sound in the time-domain.
% Fs: The sampling rate of the input signal. A typical value is 44100 Hz.
% HIGH: The maximum frequency the filter will amplify. A typical value for
% this variable is 1000 Hz.
non_stereophonic = input(:, 1); % Removes the sterophonic property of the input sound
% by just taking the first column of data.
Len = length(non_stereophonic);
F = Fs * ((-Len/2) : ((Len/2) - 1)) / Len; % Creating the array of frequencies
% which the FFT Shifted version of the signal can be plotted against.
inputFreq = fftshift(fft(non_stereophonic)); % Creates the Fourier Transform of the
% input signal. fftshift() makes it such that the zero frequency is at the
% center of the array.
lowAmplifyFilter = zeros(1, length(inputFreq));
% Creating a filter which amplifies lower frequencies.
for i = 1:length(lowAmplifyFilter)
if abs(F(i)) < HIGH
lowAmplifyFilter(i) = 1.25;
else
lowAmplifyFilter(i) = 1.00;
end
end
lowPassedInput = inputFreq .* transpose(lowAmplifyFilter); %Apply the "lowAmplifyFilter".
% Adding the chorus effect.
realOutput = real(ifft(fftshift(lowPassedInput)));
output = realOutput;
% Adding 100 randomly delayed signals to the original signal which creates the chorus effect.
for i = 1:100
currentDelay = 0.003 * rand(); % The current delay of the sound in seconds.
currentIndex = round(currentDelay * Fs); % Find the first index where the sound should start playing.
delayedOutput = [zeros(currentIndex, 1); realOutput]; % Adds "currentIndex" zeros to the front of the
% "realOutput" vector to create a slightly delayed version of the signal.
delayedOutput = delayedOutput(1:length(realOutput)); % Truncates the "delayedOutput"
% vector so that it can be added to the "realOutput" vector.
output = output + delayedOutput;
end
output = output ./ 100; % Divide by 100 to decrease the amplitude of the sound to a normal level.

56
src/muffled_effect_schluep.m Normal file
View File

@ -0,0 +1,56 @@
function output = muffled_effect_schluep(input, Fs, LOW, MED, HIGH)
% muffled_effect_schluep: Outputs a muffled version of the the original input
% sound in the time domain. This makes it sound as if the audio is being
% played in another room.
% Try this function out with "Strong_Bassline.mp3".
% CONTRIBUTORS:
% Nicolas Schluep: Function Author
% DOCUMENTATION:
% input: The input sound in the time-domain.
% Fs: The sampling rate of the input signal. A typical value is 44100 Hz.
% HIGH: The maximum frequency that the low-pass filter will let pass. A
% typical value for this variable is 1000 Hz.
non_stereophonic = input(:, 1); % Removes the sterophonic property of the input sound
% by just taking the first column of data.
Len = length(non_stereophonic);
F = Fs * ((-Len/2) : ((Len/2) - 1)) / Len; % Creating the array of frequencies
% which the FFT Shifted version of the signal can be plotted against.
inputFreq = fftshift(fft(non_stereophonic)); % Creates the Fourier Transform of the
% input signal. fftshift() makes it such that the zero frequency is at the
% center of the array.
lowPassFilter = zeros(1, length(inputFreq));
% Creating Low Pass Filter
for i = 1:length(lowPassFilter)
if abs(F(i)) < HIGH
lowPassFilter(i) = 1;
else
lowPassFilter(i) = 0;
end
end
lowPassedInput = inputFreq .* transpose(lowPassFilter); %Apply the low-pass filter.
% Adding a slight reverb effect.
realOutput = real(ifft(fftshift(lowPassedInput)));
delay = 0.001; % The delay of the sound in seconds.
index = round(delay*Fs); % Find the first index where sound should start playing
% by multiplying the delay by the sampling frequency.
delayedOutput = [zeros(index, 1); realOutput]; % Adds "index" zeros to the front of the
% "realOutput" vector to create a slightly delayed version of the signal.
delayedOutput = delayedOutput(1:length(realOutput)); % Truncates the "delayedOutput"
% vector so that it can be added to the "realOutput" vector.
output = (realOutput + delayedOutput) ./ 2.0; % Adds the "realOutput" and "delayedOutput"
% vectors to create a reverb effect. Divides by 2 to avoid clipping
% effects.

62
src/seperate_prevalent_schluep.m Normal file
View File

@ -0,0 +1,62 @@
function output = seperate_prevalent_schluep(input, Fs, LOW, MED, HIGH)
% seperate_prevalent_schluep: Attempts to seperate the most prevalent frequencies
% from the input sound by finding the most prevalent frequencies and applying
% a band-pass filter to a small region around those frequencies.
% Try this function out with "Strong_Bassline.mp3".
% CONTRIBUTORS:
% Nicolas Schluep: Function Author
% DOCUMENTATION:
% input: The input sound in the time-domain.
% Fs: The sampling rate of the input signal. A typical value is 44100 Hz.
% HIGH: The maximum distance around the maximum frequency value that will
% not be attenuated. A good range of values is usually 250-500 Hz, but it
% depends on the input sound.
non_stereophonic = input(:, 1); % Removes the sterophonic property of the input sound
% by just taking the first column of data.
Len = length(non_stereophonic);
F = Fs * ((-Len/2) : ((Len/2) - 1)) / Len; % Creating the array of frequencies
% which the FFT Shifted version of the signal can be plotted against.
inputFreq = fftshift(fft(non_stereophonic)); % Creates the Fourier Transform of the
% input signal. fftshift() makes it such that the zero frequency is at the
% center of the array.
bandPassFilter = zeros(1, length(inputFreq));
maxAmplitude = 0;
mostPrevalentFrequency = 0;
% Finding the most prevalent frequency.
for i = round(length(inputFreq)/2):length(inputFreq)
if inputFreq(i) > maxAmplitude
maxAmplitude = inputFreq(i);
mostPrevalentFrequency = F(i);
end
end
% Determining maximum and minimum frequency values for the Band Pass filter.
maxFrequency = mostPrevalentFrequency + HIGH;
minFrequency = mostPrevalentFrequency - HIGH;
if minFrequency < 0.0
minFrequency = 0.0;
end
% Creating the Band-Pass filter.
for i = 1:length(bandPassFilter)
if (abs(F(i)) < maxFrequency) && (abs(F(i)) > minFrequency)
bandPassFilter(i) = 1;
else
bandPassFilter(i) = 0;
end
end
bandPassedInput = inputFreq .* transpose(bandPassFilter); %Apply the Band-Pass Filter.
output = real(ifft(fftshift(bandPassedInput)));