Skip to content
GitLab
Projects
Groups
Snippets
Help
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Sign in
Toggle navigation
toolbox
Project overview
Project overview
Details
Activity
Releases
Repository
Repository
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Locked Files
Issues
12
Issues
12
List
Boards
Labels
Service Desk
Milestones
Iterations
Merge Requests
4
Merge Requests
4
Requirements
Requirements
List
Security & Compliance
Security & Compliance
Dependency List
License Compliance
Operations
Operations
Incidents
Analytics
Analytics
Code Review
Insights
Issue
Repository
Value Stream
Wiki
Wiki
Snippets
Snippets
Members
Members
Collapse sidebar
Close sidebar
Activity
Graph
Create a new issue
Commits
Issue Boards
Open sidebar
Institute of Technical Acoustics (ITA)
toolbox
Commits
483d6b05
Commit
483d6b05
authored
Jan 18, 2017
by
Jan-Gerrit Richter
Browse files
Options
Browse Files
Download
Email Patches
Plain Diff
added headphone equilization back
parent
c45b310e
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
82 additions
and
0 deletions
+82
-0
applications/Binaural-HRTF/HpTF_class/ita_headphone_equalization.m
...ons/Binaural-HRTF/HpTF_class/ita_headphone_equalization.m
+82
-0
No files found.
applications/Binaural-HRTF/HpTF_class/ita_headphone_equalization.m
0 → 100644
View file @
483d6b05
function
output
=
ita_headphone_equalization
(
HPTF
,
type
)
% <ITA-Toolbox>
% This file is part of the application HRTF_Measurement for the ITA-Toolbox. All rights reserved.
% You can find the license for this m-file in the application folder.
% </ITA-Toolbox>
output
=
HPTF
(
1
,
1
);
for
cdx
=
1
:
HPTF
(
1
,
1
)
.
nChannels
hp
=
merge
(
HPTF
.
ch
(
cdx
));
switch
type
case
'mean'
hp
.
freqData
=
mean
(
abs
(
hp
.
freqData
),
2
);
case
'max'
hp
.
freqData
=
max
(
abs
(
hp
.
freqData
),[],
2
);
case
'mSTD'
hp
.
freqData
=
mean
(
abs
(
hp
.
freqData
),
2
)
+
2
*
std
(
abs
(
hp
.
freqData
),
0
,
2
);
otherwise
error
(
'Unknown type'
);
end
%% Short Filter with no correction for low freqs and minimun phase
R
=
hp
;
% R = ke4;
n
=
R
.
nSamples
;
aux
=
max
(
abs
(
R
.
freqData
),[],
2
);
% find first maximum and truncate low freq correction at this point
idx1
=
find
(
R
.
freqVector
>
100
,
1
,
'first'
);
idx2
=
find
(
R
.
freqVector
<
300
,
1
,
'last'
);
d_aux
=
diff
(
aux
(
idx1
:
idx2
));
idx
=
find
(
diff
(
sign
(
d_aux
))
~=
0
,
1
,
'first'
);
aux
(
1
:
idx1
+
idx
+
1
)
=
aux
(
idx1
+
idx
+
2
);
aux
(
aux
==
0
)
=
rand
(
1
)
*
eps
;
R
.
freqData
=
aux
;
% do smoothing
%R = ita_smooth(R,'LogFreqOctave1',1/6,'Abs');
R
=
ita_invert_spk_regularization
(
R
,[
0
18000
],
'beta'
,
.
01
);
% minimum phase
R
=
ita_time_shift
(
R
,
n
/
2
,
'samples'
);
R
=
ita_extend_dat
(
R
,
R
.
nSamples
*
2
);
R
=
ita_time_shift
(
R
,
-
n
/
2
,
'samples'
);
N
=
2
*
n
;
aux
=
R
.
timeData
;
R
.
timeData
=
ifft
(
log
(
abs
(
fft
(
aux
))));
T
=
2
^
16
;
if
T
>=
N
-
3
T
=
N
-
3
;
end
if
rem
(
T
-
1
,
2
)
==
1
T
=
T
+
1
;
end
u
=
[
1
;
2
*
ones
((
N
-
T
-
1
)/
2
,
1
);
cos
(
pi
*
(
0
:
T
-
1
)
'
/
T
)
+
1
;
zeros
((
N
-
T
-
1
)/
2
,
1
)];
%pode colocar uma transicao mais suave no meio.
R
.
timeData
=
R
.
timeData
.*
u
;
H
=
R
;
H
.
timeData
=
ifft
(
exp
(
fft
(
R
.
timeData
)),
'symmetric'
);
H
=
ita_time_crop
(
H
,[
1
n
],
'samples'
);
% H = ita_time_window(H,[2^12 2^13],'samples');
% H = ita_time_window(H,[0.1 0.2],'time','dc',true);
H
=
ita_time_window
(
H
,[
2
^
9
2
^
10
],
'samples'
);
H
.
channelUnits
=
{
''
};
% It is necessary to correct the overall level of the filter
% we can either guaranty no gain
% H = H/max(abs(H.freqData));
% spl = ita_spk2level(H,3,'averaged');
%
% H = H/mean(spl.data);
output
.
freqData
(:,
cdx
)
=
H
.
freqData
;
end
% or guaranty that the average level of the signal is not altered
% this also means that the overall loudness of the signal will not be
% considerably altered.
% output = output/norm(output.rms);
\ No newline at end of file
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment