You are on page 1of 25

G.

Peeters

A Large Set of Audio Features for Sound Description

2004

 
  

!#"%$'&($')+*,"!-#./0$#&($ 12/!354(67#89:*-#!;&#<$'="0
><?7@A@BDC E& $(F0E!A/
peeters@ircam.fr
http://www.ircam.fr/

;/!$'&92G36 @DHEIJD!; &K I@@ABML

1 Introduction

1.1

Features taxonomy

NOEP0QSRUTEVUPNNXWUYZR[\VUQU]^TE_0TO [`WEabOEcUPDaSP0QUOdUYP
ecUPfOETE]gPhPiSOEP0VSOWEajOEcUPfRUPN_0YT kO'T WUVlkmYMWUn0TERUP0Rpo[XO cUPqaSP0QUOEdUYPN
UvU<w v wyx
{v
~{y { }

sK,
zD{ |A }Aw ~ ,sK2~v |
y~ <} Ux
{

rts u<vUwyx

v , wyv,wyv y
| sK,{ |
zD{y|A}A~ ~|
D xKy<wyx
sK,
zD{y|A}A~ ~|

< wyx
{yy Kx s vU~ u
{

z{y|A}A~

sK,

~|

r u v,wMs 2v
{ {
!

23/04/04

"

1/25

G. Peeters

A Large Set of Audio Features for Sound Description

2004

QUoNO YQU_0OEVUPNMN W aZOEcUPaUP0QUOdUYMP


PiSOEYQU_0OET WUVmk YWU_0PNMNgWEahO cUP2aUP0QUOEd2YP



y

   


(M U0E  
!"$#0M &%'( '  '   )*

% +*) % 

 A 0
:
 )  2

$
132 G"4 <:=

-.-0/

132 G"4 <>=8H0@ <>ABF

4 78/ K<>48K<:4 F9:587


N F>AMJ 9:@ <>=
F78C@ 2 J K9>@ 7

48F>@ G6R

4878K<:4 K<>48F9>587
N 13J FC8K @ <:=
F78C@ 2 J K9>@ 7

4878? K<>48K<>48F9:5 7
N <:@AB9:462 C
F78C@ 2 J K9:@ 7
/

F>AMJ 9:@ <:=


AT9E;6F>= 2 48G

48SF>= 9:J

$
/

= 9:P <>=
N F>AMJ 9:@ <>=
F78C@ 2 J K9>@ 7

&

= 9:P <>=
N >
F AMJ 9:@ <:=
F78C@ 2 J K9:@ 7

$
'

!
,

1.2

Organization of the paper

23/04/04

13
? 2 4658789:2 ;6<>=
<:D @AB9:462 C
9E;6F>=

2/25

F>@D C8F>J K 58<>=


9;6F>=

4 78I K<>48K<:4 F9:587


N F>@ C8F>J K 58<>=
F78C@ 2 J K9:@ 7

G. Peeters

A Large Set of Audio Features for Sound Description

2004

2 Pre-computing

2.1

Energy envelop
$

  

YM] N

2.2

    
  !   

  
     
    

Short-Time Fourier Transform

&5' (6)* +

2.3

       !" !

&' ()* +,.- * /10

2' ()* + ,.- * /1034

##%$

2' (6)* + ,B- * /103 4

,B- 0CD3 4E%* />D+ 3 4

7 )289* )8:* )0; <2


$ 0=/>; - * +
? 02@-A' >89;- 2

7 )28:* )89* )0; <2


&5>0@8- * +
? 02@ -A' >8:; - 2

Sinusoidal Harmonic modeling


^JL_:O J_:O JVMKL
Q O RAS1MJ I T
`VLTRAI ]_:M R L
X.R VY6aZ [\%O Sb]DP aZ[

H5I WJOP

23/04/04

H5I WJO PXBR O S1V

FF%G

XBR VYDZ[\.O S] P Z[

3/25

HI J KLMI NDO P


Q O RAS1MJ I T
U=MDNDV P

G. Peeters

2.4

A Large Set of Audio Features for Sound Description

2004

Perceptual model

  


  





 !"!#%$'&(
) #%$

   

!!
/ &
) 0
 #$
*+ ,

 '!,
   

-',.      ,

2.4.1 Mid-ear filtering


0

Amplitude [db20]

-20

-40

-60

-80

-100

-120
-4
10

-2

10

10

10

10

10

Frequency [Hz]

+ ,

2.4.2 Mel scale


$

Number of mel bands: 24


2
1.8
1.6
1.4
1.2
1
0.8
0.6

aS_

23/04/04

0.4

9<:

678

6 +

> ? 6

678

5 5

9;:

3 3
4 34 3

0.2
0

0.5

1.5

Frequency [Hz]

2.5
4

x 10

/ ,

4/25

G. Peeters

A Large Set of Audio Features for Sound Description



   
!#"%$&'




=


(!)


 
=    

oUP:TEV

 


P0VUR 

2004

 




2.4.3 Bark scale


$
Number of bark bands: 24
1
0.9
0.8
0.7
0.6
0.5
0.4

-./

,
0
+

-./

,
0
+

0.3

0.2
0.1
0

0.5

1.5

Frequency [Hz]

0 1



5768  4 
1?!@#ACB1DE
( )

23/04/04

>
=

=)

2.5
x 10

< )

9
 
= 75 :; 6 4

5/25

G. Peeters

2.5

A Large Set of Audio Features for Sound Description

2004

Amplitude and Frequency scale

2.5.1 Amplitude scales


0.1

Ampl

Ampl

0.1

0.05

0
2000
4000
-3
Freq
x 10

0.05

0
-10
-3
x 10
8

6000

2.5.2 Frequency scales

2000

0
-10

6000

Freq

100
50
0

2000

-5

4000

50
0
-10

6000

-5
Log-freq

4
4

(
6

.
5

5 5

.
.

5
5

6
-

.
.

(
(
6

(
-

Descriptors on Spectrum / Harmonic peaks / Bark bands

23/04/04

100

3 &
4

2.6

150

Freq

Log-freq

200
Log-ampl

Log-ampl

4000

150

200

2
0

0
Log-freq

6
Power

Power

-5

6/25

G. Peeters

A Large Set of Audio Features for Sound Description

2004

3 Global temporal features


3.1

Envelop characterization

3.1.1 Attack / Decay / Sustain / Release envelop modeling

attack decay

7 %

sustain

release

& 9

( "
!

sustained sound

non-sustained sound

attack

: %

rest

( "
!

23/04/04

7/25

G. Peeters

A Large Set of Audio Features for Sound Description

2004

3.1.2 Attack part

3.1.2.1 Estimation of the start and end of the attack


!

energy

attack
end

start

90%

...
20%

time

"

$
energy

O;
O; O;


...
threshold 2
threshold 1

OEc



@ 

OEc
;

56 8

OEc



time

23/04/04

end

effort 12

OEc

start

effort 23
...

OEc

attack

OEc



O 56 8

56 8

8/25

G. Peeters

A Large Set of Audio Features for Sound Description

3.1.2.2 Log-Attack Time

(mpeg7:LogAttackTime)

2004

DT.g_lat



3.1.2.3 Temporal increase

(cuidado:TemporalIncrease) DT.g_incr
;

3.1.3 Sustain part

3.1.3.1 Decrease part: Temporal decrease (cuidado:TemporalDecrease) DT.g_decr


$


<



>

3.1.3.2 Sustain part: Energy Modulation and Fundamental frequency modulation


(mpeg7:AudioPower ScalableSeriesType element name="Modulation")
(mpeg7:AudioFUndamentalFrequency ScalableSeriesType element name="Modulation")

23/04/04

9/25

G. Peeters

A Large Set of Audio Features for Sound Description

2004

3.1.4 Example
F:\data\class\sol\sust\bowedstring\alto\mf\alto\_a\_gref\_mf\_si3\_12.wav
1

0.5

0.2

0.4

0.6
0.8
1
1.2
1.4
Dlat: -0.53981 - threshold: 0.15 - Dincr: 3.265 - Ddecr: -0.28535

1.6

1.8

10

1.4

1.6

1.8

15000
10000
5000
0

4
5
6
incr (r-) incr2(r--) desc (r-)

0.5

satt_posn

0.2

0.4

; <
4#

5 %

4,

eatt_posnmaxenv_posn

0.6

2 #

1.2

"

5 %
5 9

41

0.8

2 (

0.4
0.3
0.2
0.1
0

0.5

1
1.5
MODam: 0.060872 - MODfr: 5.3833

2.5

0.6
envelop-v
polyfit
hatenvelop-v

0.4
0.2
0
-0.2
0.6

0.8

1.2

1.4

1.6

1.8

2.2

0.015
fft(envelopv-polyfit)
0.01
0.005
0

4,
41

23/04/04

5 %

10

= %
4#
5 %
% 8

15

20

25

30

35

40

10/25

45

50

G. Peeters

3.2

A Large Set of Audio Features for Sound Description

2004

Others

3.2.1 Temporal centroid

(mpeg7:TemporalCentroid)

DT.g_tc

(cuidado:TemporalEffectiveDuration)

DT.g_ed






3.2.2 Effective Duration

energy

threshold

time

effective duration

4 Instantaneous temporal features


4.1

Auto-correlation

(cuidado:AudioZcr)

DT.i_xcorr_m

$
0.2

1
signal

xcorr
0.5

Amplitude

Amplitude

0.1
0

-0.1
-0.2

200

400
Time

600

-0.5
-20

800

-10

0
Time

10

20

250
signal
xcorr

Amplitude

200
150
100
50
0



1000

2000

4
4

3000
Frequency

4000

5000

6000

5
8

23/04/04

11/25

G. Peeters

4.2

A Large Set of Audio Features for Sound Description

Zero-crossing rate (cuidado:AudioXcorr)

2004

DT.i_zcr_v

' >
8

+ >

? / + '

? 7 0 =

5 Energy features
5.1

Total Energy

(mpeg7:AudioPower)

DE.i_tot_v

5.2

Harmonic Part Energy

(cuidado:AudioHarmonicPower)

DE.i_harmo_v

(cuidado:AudioNoisePower)

DE.i_noise_v

5.3

Noise Part Energy


$

23/04/04

12/25

G. Peeters

A Large Set of Audio Features for Sound Description

2004

6 Spectral features
6.1

Spectral shape description

6.1.1 Spectral centroid

(mpeg7:AudioSpectrumCentroid)

 


6.1.2 Spectral spread

 =

DS.i_sc_v

 

  
 

(mpeg7:AudioSpectrumSpread)

DS.i_ss_v

6.1.3 Spectral skewness (cuidado:AudioSpectrumSkewness)

DS.i_skew_v
m e a n : 7 . 8 7 2 e - 0 1 7 s td : 5 s k e w : - 8 . 3 2 5 4 e - 0 1 7 k u r t: 3
d a ta
ga u s s f it

0.0 9





 

0.0 8

0.0 7

0.0 6

0.0 5

 =  

0.0 4

0.0 3

0.0 2

0.0 1

0
- 50

- 40

-30

-2 0

-1 0

10

20

30

40

50

m e a n : 1 6 .6 7 s td : 2 3 . 5 7 1 4 s k e w : - 0 . 5 6 5 6 9 k u r t: 2 . 4
0 .0 2 5

d a ta
g a u s s fi t

0 .0 2

0 .0 1 5

0 .0 1

0 .0 0 5

0
-5 0

-4 0

-3 0

-2 0

-1 0

10

20

30

40

50

m e a n : - 1 6 .6 7 s t d : 2 3 . 5 7 1 4 s k e w : 0 . 5 6 5 6 9 k u r t : 2 . 4
0 .0 2 5
d a ta
g a u s s fi t

0 .0 2

0 .0 1 5

0 .0 1

0 .0 0 5

0
-5 0

23/04/04

-4 0

-3 0

-2 0

-1 0

10

20

30

40

50

13/25

G. Peeters

A Large Set of Audio Features for Sound Description

6.1.4 Spectral kurtosis

(cuidado:AudioSpectrumKurtosis)

DS.i_kurto_v

m e a n : 7 . 8 7 2 e -0 1 7 s t d : 5 s k e w : - 8 . 3 2 5 4 e - 0 1 7 k u rt : 3
d a ta
g a u s s f it

0 .0 9

 =

2004

0 .0 8

0 .0 7
0 .0 6
0 .0 5

0 .0 4

 =  

0 .0 3
0 .0 2
0 .0 1
0
-5 0

-4 0

-3 0

-2 0

-1 0

10

20

30

40

50

m e a n : - 2 . 1 5 9 7 e -0 1 5 s td : 2 8 . 8 7 0 4 s k e w : 3 . 1 2 0 4 e - 0 1 6 k u r t: 1 .8
0 .0 2 5
d a ta
g a us s fi t

0 .0 2

0 .0 1 5

0 .0 1

0 .0 0 5

-1 0

-8

-6

-4

-2

10

m e a n : 0 . 0 0 4 9 9 9 8 s t d : 1 . 4 1 4 2 s k e w : 5 .3 0 3 2 e - 0 0 7 k u r t: 6 .0 0 0 3
1 .2

d a ta
g a u s s f it

0 .8

0 .6

0 .4

0 .2

0
-1 0

6.1.5 Spectral slope

(cuidado:AudioSpectrumSlope)

-8

-6

-4

-2

10

DS.i_slope_v

$ 
=    + 

 

 "! = 


23/04/04

 

 


 


#





 



#

 

 

14/25

G. Peeters

A Large Set of Audio Features for Sound Description

6.1.6 Spectral decrease (cuidado:AudioSpectrumDecrease)

2004

DS.i_decr_v

   

 =



=







6.1.7 Spectral roll-off

(cuidado:AudioSpectrumRollOff)

DS.i_rolloff_v





 

 




NY



"

8
.

6.2

"

/ 4#
5 %
; 0 @
4
5
; 0 @
8
-

.
-

8
.

Temporal variation of spectrum

6.2.1 Temporal variation of spectrum: spectral variation (cuidado:AudioSpectrumVariation)


DS.i_var_v
$

Q 
O
$

Q
O






=



23/04/04

 







15/25

G. Peeters

6.3

A Large Set of Audio Features for Sound Description

2004

Global spectral shape description

6.3.1 Mel Frequency Cepstral Coefficients (MFCC) (cuidado:AudioMFCC) DP.i_MFCC_m


$
Log-amplitude

-5
spectrum
mid-ear spectrum

-10
-15
-20

1000

2000

3000
Frequency

Log-amplitude

4000

5000

6000

Mel band spectrum


MFCC spectrum

0
-1
-2
-3

10

15

20

25

Mel band

MFCC
Value

$
s(n)

FFT

MelBand

0
-5
-10

0 4#

MFCC

DCT

6
MFC coefficient

5 ,

Log

,
4

-,

 
  

-,




= 



5 ,

 

7 Harmonic features
7.1.1 Fundamental frequency

(mpeg7:AudioFundamentalFrequency) DH.i_f0_v

7.1.2 Noisiness

(mpeg7:AudioHarmonicity)

23/04/04

10

DH.i_noisiness_v

   
 


16/25

12

G. Peeters

A Large Set of Audio Features for Sound Description

(cuidado:AudioInharmonicity)

    


f0

energy

7.1.3 Inharmonicity

2004

DH.i_inharmo_v

2 f0 3 f0 4 f0 5 f0 6 f0 7 f0

f(1)

f(2) f(3) f(4) f(5) f(6)


frequency

$
(

7.1.4 Harmonic Spectral Deviation

(mpeg7:HarmonicSpectralDeviation) DH.i_devs_v
Spectral deviation: 0.15374
0.18
spectral envelop
harmonics

0.16

(







c


Q 
c

c
a 
c

0.14

0.12
Amplitude



0.1
0.08
0.06

0.04
0.02
0

4
6
Frequency [harm number]

10

7 )
8

23/04/04

17/25

G. Peeters

A Large Set of Audio Features for Sound Description

2004

7.1.5 Odd to Even Harmonic Energy Ratio (cuidado:HarmonicSpectralOERatio):


DH.i_oeratio_v
$

Odd/even harmonic energy ratio: 3.2431


0.045
odd harmonic
even harmonic

0.04
0.035

Amplitude

0.03

  

 



0.01
0.005

  
=

0.02
0.015

0.025

10
Frequency [harmonic number]

15

20

% 8

7.1.6 Tristimulus (cuidado:HarmonicSpectralTristimulus): DH.i_tri*_v


$

tri1: 0.49442 tri2: 0.45368 tri3: 0.0519


0.4

P0ODQ ,

tristimulus1
tristimulus2
tristimulus3

0.35
0.3
0.25
0.2
0.15

$
0.1

+





0.05




23/04/04

+

10

15

20




18/25

G. Peeters

A Large Set of Audio Features for Sound Description

2004

8 Perceptual features
8.1

Features

8.1.1 Total Loudness and specific loudness (cuidado:AudioLoudness): DP.i_loud_v


$

  
  

=
      


 =
   

8.1.2 Relative Specific Loudness


DP.i_specloudnorm_m

 

  

8.1.3 Sharpness

       






  =   <
 =

(cuidado:AudioSharpness)

DP.i_sharp_v

8.1.4 Spread

)+*

(cuidado:AudioRelativeSpecificLoudness):

$%& '

23/04/04





(cuidado:AudioSpread)

DP.i_spread_v

,' ' ( #" !

19/25



G. Peeters

A Large Set of Audio Features for Sound Description

2004

9 Various features
9.1

Spectral Flatness/Crest measure

(mpeg7:AudioSpectrumFlatness) DP.sfm_m

&

 ! 





 



 




 




 
 

&

13254

, *-.

-/0

(* ()

&
9;<>=?
FHGIJKMLON P

$&'
6

"#$% )

=
=

78:9;<
CDE

9;<

B
@

=?
A

$
QSR

23/04/04

))

*
+

, *-.

-/0

Z\[^]

UWV


T

_a`

Y
X

_a`

20/25

G. Peeters

A Large Set of Audio Features for Sound Description

2004

10 Temporal modeling

10.1.1 Mean


  


  


 

 




10.1.2 Variance
   
*

, "!$#%'&

   
(

, "!$#%'&

) , "!$#%'&

10.1.3 Deviation
/ . 0 1243
A;6 BDC

E
=

H 5?6 8$:@;

576 8:<;
/ . 0 1243
=

(>

H 57698$:<;
G

H 576 8:<;

57698$:<;

E
F

+
+

>

H 576 8:<;

10.1.4 Temporal modeling an mpeg-7 audio scalable series


mpeg7::scalableseries.
scalableseries
AudioLoudnessType
mpeg7:scalableseries
numOfElements=1

Element Name

weight
element name

Mpeg-7
yes

Mean

Yes
Variance

Extension
Derivative

extension
Modulation

23/04/04

21/25

G. Peeters

A Large Set of Audio Features for Sound Description

#$!  %& '( )*

#$!  %& '( )*


, .-

34 375 3698:69;75 3


<0=7E;F=

34 35 37698:69;75 3


<0=7>&3;769=

/    

 
  
 
   

2004

   
 !   "

34 35 37698:69;75 3


<0=?0;7@ A ;6CBD37=

!  / & /

34 35 37698:69;75 3


<0=7G93@ A H;78 A H3=

, !  0 -

34 35 37698:69;75 3


<0=7>&I7J9K94 ;78 A I769=

12121
+ " , & -

' = #

23/04/04

"

22/25

G. Peeters

A Large Set of Audio Features for Sound Description

2004

11 List of all descriptors


LLD List
Temporal Features
Global Temporal Features
Log Attack Time
Temporal Increase
Temporal Decrease
Temporal Centroid
Effective Duration
Instantaneous Temporal Features
Signal Auto-correlation function
Zero-corssing rate
Energy Features
Total energy
Total energy Modulation (frequency, amplitude)
Total harmonic energy
Total noise energy
Spectral Features
Spectral Shape
Spectral centroid
Spectral spread
Spectral skewness
Spectral kurtosis
Spectral slope
Spectral decrease
Spectral rolloff
Spectral variation
Global spectral shape description
MFCC
Delta MFCC
Delta Delta MFCC
Harmonic Features
Fundamental frequency
Fundamental fr. Modulation (frequency, amplitude)
Noisiness
Inharmonicity
Harmonic Spectral Deviation
Odd to Even Harmonic Ratio
Harmonic Tristimulus
Harmonic Spectral Shape
HarmonicSpectral centroid
HarmonicSpectral spread
HarmonicSpectral skewness
HarmonicSpectral kurtosis
HarmonicSpectral slope
HarmonicSpectral decrease
HarmonicSpectral rolloff
HarmonicSpectral variation
Perceptual Features
Loudness
RelaitveSpecific Loudness
Sharpness
Spread
Perceptual Spectral Envelope Shape
Perceptual Spectral centroid
Perceptual Spectral spread
Perceptual Spectral skewness
Perceptual Spectral kurtosis
Perceptual Spectral Slope
Perceptual Spectral Decrease
Perceptual Spectral Rolloff
Perceptual Spectral Variation
Odd to Even Band Ratio
Band Spectral Deviation
Band Tristimulus
Various features
Spectral flatness
Spectral crest
Total Number of Features

23/04/04

frame
based

number of
features

acronym

xml tag

n
n
n
n
n

1
1
1
1
1

DTg_lat
DTg_incr
DTg_decr
DTg_tc
DTg_ed

mpeg7:LogAttackTime
cuidado:TemporalIncrease
cuidado:TemporalDecrease
mpeg7:TemporalCentroid
cuidado::TemporalEffectiveDuration

y
y

12
1

DTi_xcorr_m
DTi_zcr

cuidado:AudioXcorr
cuidado:AudioZcr

y
n
y
y

1
2
1
1

DEi_tot_v
DTg_mod_fr, DTg_mod_am
DEi_harmo_v
DEi_noise_v

mpeg7:AudioPower
ScalableSeriesType element name="Modulation"
cuidado:AudioHarmonicPower
cuidado:AudioNoisePower

y
y
y
y
y
y
y
y

6
6
6
6
6
1
1
3

DSi_sc_m
DSi_ss_m
Dsi_skew_m
Dsi_kurto_v
Dsi_slope_v
Dsi_decs_c
Dsi_rolloff_v
Dsi_variation_v

mpeg7:AudioSpectrumCentroid (mpeg7:SpectralCentroid)
mpeg7:AudioSpectrumSpread
cuidado:AudioSpectrumSkewness
cuidado:AudioSpectrumKurtosis
cuidado:AudioSpectrumSlope
cuidado:AudioSpectrumDecrease
cuidado:AudioSpectrumRollOff
cuidado:AudioSpectrumVariation

y
y (post)
y (post)

12
12
12

DPi_mfcc_m
DPi_Dmfcc_m
DPi_DDmfcc_m

cuidado:AudioMFCC

y
n
y
y
y
y
y

1
2
1
1
3
3
9

DHi_f0_v
F0 Mod AM, FR
DHi_noisiness_v
DHi_inharmo_v
DHi_devs_v
Dhi_oeratio_v
Dhi_tri_v

mpeg7:AudioFundamentalFrequency
ScalableSeriesType element name="Modulation"
mpeg7:AudioHarmonicity
cuidado:AudioInharmonicity
mpeg7:HarmonicSpectralDeviation
cuidado:HarmonicSpectralOERatio
cuidado:HarmonicSpectralTristimulus

y
y
y
y
y
y
y
y

6
6
6
6
6
1
1
3

DHi_sc_m
DHi_ss_m
DHi_skew_m
DHi_kurto_v
DHi_slope_v
DHi_decs_c
DHi_rolloff_v
DHi_variation_v

mpeg7:HarmonicSpectralCentroid
mpeg7:HarmonicSpectralSpread
cuidado:HarmonicSpectralSkewness
cuidado:HarmonicSpectralKurtosis
cuidado:HarmonicSpectralSlope
cuidado:HarmonicSpectralDecrease
cuidado:HarmonicSpectralRollOff
mpeg7:HarmonicSpectralVariation

y
y
y
y

1
24
1
1

DPi_loud_v
DPi_specloud_m
DPi_sharp_v
DPi_spread_v

AudioLoudness
cuidado:AudioRelativeSpecificLoudness
cuidado:AudioSharpness
cuidado:AudioSpread

y
y
y
y
y
y
y
y
y
y
y

6
6
6
6
6
1
1
3
3
3
9

DPi_sc_m
DPi_ss_m
DPi_skew_m
DPi_kurto_v
DPi_slope_v
DPi_decs_c
DPi_rolloff_v
DPi_variation_v
DP_ioeratio_v
DPi_devs_v
DPi_tri_v

cuidado:AudioFilterbankCentroid
cuidado:AudioFilterbankSpread
cuidado:AudioFilterbandSkewness
cuidado:AudioFilterbankKurtosis
cuidado:AudioFilterbankSlope
cuidado:AudioFilterbankDecrease
cuidado:AudioFilterbankRolloff
cuidado:AudioFilterbankVariation
cuidado:AudioFilterbankOERatio
cuidado:AudioFilterbankDeviation
cuidado:AudioFilterbankTristimulus

y
y

4
4
166

DPi_sfm_m
DPi_scm_m

mpeg7:AudioSpectrumFlatness
cuidado:AudioSpectrumCrest

23/25

G. Peeters

A Large Set of Audio Features for Sound Description

2004

12 Acknowledgement

13 References


  



   % +
  % 3 >7 % 


0


# '    % )   %  %  %    %

>0

A % 

3


  

  D '(  ! % )    %  % A %   0  '   % 3 %  #  '      %
A E +A 0
   = 0 



 


0( %  %   \ 2
> 



A  
> %  % 

 

'&

0 % 


'

    7 M0( %  %   




  0

0

% 

 

  3 U  !  % T+" # #$t


-  %* E  > % .* ! 

3 

'

3 

 


'

. .  

/:
 %'

'

% 

7

 


H

2"*

 

'

 )" '

'  )   % 

   :E
.2*    f7 % '   % 
   

2L$

HI&"JKK
;.$ 

%'

0

#

3.%

  )


 %

% 

 HS6  )UTIH

 .   .VJ

% '

p 0
tA #0" 3" .  @ ;
/  0 > 
 J J"JK
/ 0
p -/P"@C"JKK J; 0

ZM _J     Z  !
/ `<


 % (0   % 54

>)2

' 




3 %'

 '

%

% % 20>)

A3 

  )

E

%('>% F

*) %

    %

D *# %' +)

% E

 ' 

"% 

0' ' 

 



.3 

 



 

# '    % )    %  %  % 




'& 

'!   (>  %  % 

 

 "


p>)E02^ 2
)

E0 ' A  %   $E  > >2 ,P * % ' )




a .R 
" J KK J

 %   0 L$  
2> ) ?]t>*)   

% 



' 7M 3A>7 % 2" M 

 3


>3 .Q UpM

D  '     .M %  % 

%

U
  S S-c
 '   
U
 US
 )  

)* '

 

McbJ KKK
 0

7 '    7>

M. %   % 

 ))

 Y"JK KJ ;   '  %    %  f


  % 

!
03,  
5

0

# % 

 9]

 )




"  !;
#&<   )" '

23/04/04

Z

  )

'

0


 % U8MA % 7
.W."T

 %   0 h
>#>7

  "

U+
%

") %

 



 - >9

7 %  %  % 

 )&

01%* 2
   % &b+
'

Z/ ^

;$


7 %  % M
c %  

 L

M

 %


q0#00*

  L

 %   E 0y

3 %'&%Ad 




") %

%   


y    ' &3 ^0,p -/P"@

 T)

    0  M^


M> %  % 

2 % 
N/   )7h      7 .  @
 D   7 #* % 
-   "% *y! '   X



%

.M % 0


 E0=>

  )S6!8 >)

U
 tS:  )76I98\>):"JK K ;g+ % 3 %   7 %   $t
U %  
] A 
    % +  0 7 '    M3#>7   <-6]t  )  * HS
8M

Z  ( % 3:]t>*)   7 U

  )

U
 USVbJ KK  0 '   % 
7 %  % A %   0N] P   % 3 0 #  '     3#>A
  %
  V3 %'&%ed  %    /U
R D 9'    %    W.fT01%" I 
 !
 % 8MA % ?

 )" %% .:WS  : ' % " MMX


#  
;Y4  % )* %   0 ' " % ) % '    %   3M) % 
    ' >)DE0  M   >3$) % 7 %'&%   %  % M
   '   % +y %' #0

 R !pM  > A0 '  % Z[ .Q0 0 "D  M & >7 % .[+A \  %    * ! 
%

 '

 

"% 


% )")

'

20E N O-  )" ' t M .   


 3(0 %  A %   )*h % ' 0# 
 ) 7   
 %  %  %  &  +y  2 0 % 2"  P>0  )   2 A % 2"  / QJR"QJ 
 %  

%('

 

,+

%"   > %

>)M&p0 
  
'

  2 %   %

 )*

7  0 
, M  % .*

   3

  )" '

  A #>A)B8  %  ' 30C   %  




 )>

)*) % %('

%'& 


 %

 %   ) 
  E10.2*     E


"  )76!98 >) M:



 ;9<
  )=+>p   -  

<U 0,?; .  @ ; D


U
>+) %    050  G/ 

h0:  
 % +' 0

&% % 

M

A %  %  %   0 




 M

# E  0
h  +9Z '   % &) % . M % '

* % 

24/25

A0


% 

G. Peeters

A Large Set of Audio Features for Sound Description

(?6I$  )f
   % ;

D


> )*a/ U$   ' hM MG



 

V< 0([ >P8M   <-]
'&   % 

6SV8\>) 2 
$.bJ KK .;, 2
+-/
 % .%  /V # @@#&M y
>M)"



 % .%  /V # K#M S7 32>




 % .%  /  


E
.2*     P


23/04/04

M % &


3 :

A %  %  %   ^7M  


2
 $ J K@"JJQ

M 

L8M

hh+


# %  ' 30

M) M. %   


!) #>A; *%(' 

 *)

% >

]t '

3") %





  %  M %


 >


R J
#RJ R

[ 

' % #>'M

L8M

   %     + '

 3

+

<-6

0% 0

 )"#

  

 

 +)

%

  

* % '

 %  !1"4!3

'  %   .3 


 7 %   E 0M %  % ( #>  )    )&! %  % B#> 

 3

E0!3+"  %  >")  M 7 0q '( 0  ' A %  #> 7>  ) P

:*)0-


K#&

)D/V$

b#  7 0 '   %    #>7+) 

 3)"`
#  
;= >  )

2004

)

  0   %   0 

% ) & M

25/25

You might also like