You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
74 lines
3.1 KiB
74 lines
3.1 KiB
function AACSeq2 = AACoder2(fNameIn)
|
|
%Implementation of AAC encoder
|
|
% Usage AACSeq2 = AACoder2(fNameIn), where:
|
|
% Inputs
|
|
% - fNameIn is the filename and path of the file to encode
|
|
%
|
|
% Output
|
|
% - AACSeq2 is an array of structs containing K structs, where K is
|
|
% the number of computed frames. Every struct of the array consists
|
|
% of a frameType, a winType, chl.TNScoeffs which are the quantized TNS
|
|
% coefficients of this frame's left channel, chr.TNScoeffs which are
|
|
% the quantized TNS coefficients of this frame's right channel,
|
|
% chl.frameF which are the MDCT coefficients of this frame's left
|
|
% channel, chr.frameF which are the MDCT coefficients of this frame's
|
|
% right channel
|
|
|
|
% Declares constant window type
|
|
WINDOW_TYPE = 'KBD';
|
|
|
|
% Reads the audio file
|
|
[originalAudioData, ~] = audioread(fNameIn);
|
|
|
|
% Splits the audio in frames and determines the type of each frame
|
|
frameTypes{ceil(length(originalAudioData) / 1024), 1} = 'OLS';
|
|
frameTypes{1} = 'OLS';
|
|
for i = 1:length(frameTypes) - 2
|
|
nextFrameStart = (i + 1) * 1024 + 1;
|
|
nextFrameStop = nextFrameStart + 2047;
|
|
if nextFrameStop > length(originalAudioData)
|
|
tmpZeroPadded = originalAudioData(nextFrameStart:length(originalAudioData), :);
|
|
tmpZeroPadded = padarray(tmpZeroPadded, [(2048 - length(tmpZeroPadded)) 0], 'post');
|
|
frameTypes{i+1} = SSC(1, tmpZeroPadded, frameTypes{i});
|
|
|
|
clearvars tmpZeroPadded
|
|
else
|
|
frameTypes{i+1} = SSC(1, originalAudioData(nextFrameStart:nextFrameStop, :), frameTypes{i});
|
|
end
|
|
end
|
|
|
|
% Assigns a type to the last frame
|
|
if strcmp(frameTypes{length(frameTypes) - 1}, 'LSS')
|
|
frameTypes{length(frameTypes)} = 'ESH';
|
|
elseif strcmp(frameTypes{length(frameTypes) - 1}, 'ESH')
|
|
frameTypes{length(frameTypes)} = 'ESH';
|
|
else
|
|
frameTypes{length(frameTypes)} = 'OLS';
|
|
end
|
|
|
|
% Encodes audio file
|
|
AACSeq2(length(frameTypes), 1) = struct;
|
|
for i = 0:length(frameTypes) - 1
|
|
currFrameStart = i * 1024 + 1;
|
|
currFrameStop = currFrameStart + 2047;
|
|
if currFrameStop > length(originalAudioData)
|
|
tmpZeroPadded = originalAudioData(nextFrameStart:length(originalAudioData), :);
|
|
tmpZeroPadded = padarray(tmpZeroPadded, [(2048 - length(tmpZeroPadded)) 0], 'post');
|
|
frameF = filterbank(tmpZeroPadded, frameTypes{i+1}, WINDOW_TYPE);
|
|
|
|
clearvars tmpZeroPadded
|
|
else
|
|
frameF = filterbank(originalAudioData(currFrameStart:currFrameStop, :), frameTypes{i+1}, WINDOW_TYPE);
|
|
end
|
|
|
|
[frameF(:, 1), TNScoeffsL] = TNS(frameF(:, 1), frameTypes{i+1});
|
|
[frameF(:, 2), TNScoeffsR] = TNS(frameF(:, 2), frameTypes{i+1});
|
|
|
|
AACSeq2(i + 1).frameType = frameTypes(i + 1);
|
|
AACSeq2(i + 1).winType = WINDOW_TYPE;
|
|
AACSeq2(i + 1).chl.TNScoeffs = TNScoeffsL;
|
|
AACSeq2(i + 1).chr.TNScoeffs = TNScoeffsR;
|
|
AACSeq2(i + 1).chl.frameF = frameF(:, 1);
|
|
AACSeq2(i + 1).chr.frameF = frameF(:, 2);
|
|
end
|
|
end
|
|
|