A large set of audio features for sound description ... - WWW Ircam

G. Peeters A Large Set of Audio Features for Sound Description 2004
1 Introduction
1.1 Features taxonomy
¦¤ ¤
¢¡¤£¦¥§¥©¨£¢¡¤¡¤¡¤¨
§¤¤¤¤
©
peeters@ircam.fr
http://www.ircam.fr/
¤¤ ¤¤©¤
¦¤ ¢©©¤
•
•
¤
¦
23/04/04 1/25
¤
! "

G. Peeters A Large Set of Audio Features for Sound Description 2004
•
•
•
•
¦©¤
¡
©
¤© ¤©¤
¢¤£ ¦¥¡¨§© ¢ ¨£¡¤¨¦¡ ¡¤£¥¥¡¨¡¤¥¡¡ £ ¡ ¡ £ ¡ ¦£ £ ¥©¡¨¡¤¨¡¤¡ £ ¥¡
• # $
• #
• % $
• & $
• ) $
• * $
•
1.2 Organization of the paper
23/04/04 2/25
¦
¦
¦
¦
¦
' ( ( (
!
¦
¦

G. Peeters A Large Set of Audio Features for Sound Description 2004
2 Pre-computing
•
•
•
•
2.1 Energy envelop
$
2.2 Short-Time Fourier Transform
2.3 Sinusoidal Harmonic modeling
23/04/04 3/25
¢¡ £¥¤§¦©¨
§©
¥§¥
¤ £
¤ ¨
¤ £ ¤§
¢ ¥§©§ © ¥§© ©
§©§©§©§
§© ©
§ §¥
©¦©¨ ¨
¥ ¦©¨ ©
¡ ©
§© © ©
¥§©§ ©
©
§§©§©§©§
© §
©
§©§©§¥§
©¥
©
§© §
§¥ © ¢
©¥
©

G. Peeters A Large Set of Audio Features for Sound Description 2004
2.4 Perceptual model
•
•
2.4.1 Mid-ear filtering
2.4.2 Mel scale
$
• =
• = ⋅
+
£¢ ¤¦¥§©¨ ¡
¨ ¨¨¨
23/04/04 4/25
¤¦¥§©¨
Amplitude [db20]
0
-20
-40
-60
-80
-100
10 -4
-120
2
1.8
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
10 -2
10 0
10 2
Frequency [Hz]
10 4
+ , - .
Number of mel bands: 24
¥
0 0.5 1 1.5 2 2.5
x 10 4
0
Frequency [Hz]
/ ,
10 6

G. Peeters A Large Set of Audio Features for Sound Description 2004
=
2.4.3 Bark scale
=
⋅
$
=
¡
§
§
+
=
=
§
§
⋅
¡
¡ £¢¥¤¡§¦©¨ ¨ ¤¡¦
0 0.5 1 1.5 2 2.5
x 10 4
0
Frequency [Hz]
23/04/04 5/25
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
Number of bark bands: 24
0 1 2
¡

G. Peeters A Large Set of Audio Features for Sound Description 2004
2.5 Amplitude and Frequency scale
2.5.1 Amplitude scales
•
•
•
2.5.2 Frequency scales
•
•
0
0
x 10-3
8
2000 4000
Freq
6000
23/04/04 6/25
Ampl
Log-ampl
0.1
0.05
Power
6
4
2
0
0 2000 4000 6000
Freq
200
150
100
50
0
0 2000 4000 6000
Freq
Ampl
Log-ampl
0.1
0.05
Power
0
-10
x 10-3
8
-5 0
Log-freq
5
6
4
2
0
-10 -5 0 5
Log-freq
200
150
100
3 &
4 - 5 - . 6 - (
4 - 5 - . 6 (
4 - 5 - . 6 - (
4 - 5 - . 6 - (
4 - 5 - . 6 (
4 - 5 5 - . 6 -
2.6 Descriptors on Spectrum / Harmonic peaks / Bark bands
50
0
-10 -5 0 5
Log-freq

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
7
:
attack decay sustain release
7 % 8 $ & 9 8 ( "
!
sustained sound
non-sustained sound
attack rest
: % 8 $ 9 8 ( "
!
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
! $
8 " 2
$
¢¡¤£¤¥
§¦¢¨
¦¢¨
©
©
¦¢¨
23/04/04 8/25
90%
...
20%
energy
energy
...
threshold 2
threshold 1
effort 12
start
attack
attack
start end
effort 23
...
end
time
time

G. Peeters A Large Set of Audio Features for Sound Description 2004
3.1.2.2 Log-Attack Time (mpeg7:LogAttackTime) 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
$
α
= ⋅ −α
−
¨©
>
¨©
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
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Dlat: -0.53981 - threshold: 0.15 - Dincr: 3.265 - Ddecr: -0.28535
15000
23/04/04 10/25
1
0.5
10000
5000
F:\data\class\sol\sust\bowedstring\alto\mf\alto\_a\_gref\_mf\_si3\_12.wav
0
0 1 2 3 4 5 6 7 8 9 10
incr (r-) incr2(r--) desc (r-)
1
0.5
satt_posn eatt_posnmaxenv_posn
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
; < - 2 # # 6
4# 5 % % 8 " 8
4, 5 %
41 5 9 2 (
0.4
0.3
0.2
0.1
0
0 0.5 1 1.5 2 2.5
MODam: 0.060872 - MODfr: 5.3833
0.6
0.4
0.2
0
-0.2
0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2
0.015
0.01
0.005
envelop-v
polyfit
hatenvelop-v
fft(envelop v -polyfit)
0
0 5 10 15 20 25 30 35 40 45 50
= %
4# 5 % % 8
4, 5 % % 8
41 5 8

G. Peeters A Large Set of Audio Features for Sound Description 2004
3.2 Others
3.2.1 Temporal centroid (mpeg7:TemporalCentroid) DT.g_tc
$
¡¤£
=
$
¢
¢
¡
¡
⋅
¡
3.2.2 Effective Duration (cuidado:TemporalEffectiveDuration) DT.g_ed
$
4 Instantaneous temporal features
threshold
4.1 Auto-correlation (cuidado:AudioZcr) DT.i_xcorr_m
£
=
$
$
§
©
− −
¨
=
¦
¥
⋅
+
23/04/04 11/25
Amplitude
Amplitude
0.2
0.1
0
-0.1
energy
-0.2
0 200 400
Time
600 800
250
200
150
100
50
effective duration
1
signal xcorr
Amplitude
0.5
0
-0.5
-20 -10 0
Time
10 20
0
0 1000 2000 3000
Frequency
4000 5000 6000
4 5
8 -
signal
xcorr
4 - 5 4 - 5 -
time

G. Peeters A Large Set of Audio Features for Sound Description 2004
4.2 Zero-crossing rate (cuidado:AudioXcorr) DT.i_zcr_v
$
5 Energy features
' > - ? / + '
8
+ > - ? 7 0 =
8
5.1 Total Energy (mpeg7:AudioPower) DE.i_tot_v
5.2 Harmonic Part Energy (cuidado:AudioHarmonicPower) DE.i_harmo_v
$
5.3 Noise Part Energy (cuidado:AudioNoisePower) DE.i_noise_v
$
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) DS.i_sc_v
μ =
•
⋅
¡
δ
£¥¤§¦©¨ ¢
• ¡ =
=
¡
¡
6.1.2 Spectral spread (mpeg7:AudioSpectrumSpread) DS.i_ss_v
σ = − μ
⋅
¡
δ
6.1.3 Spectral skewness (cuidado:AudioSpectrumSkewness) DS.i_skew_v
= − μ ⋅ δ
•
•
•
•
•
¡
=
γ
σ
23/04/04 13/25
¢
¢
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 . 0 2 5
0 . 0 2
0 . 0 1 5
0 . 0 1
0 . 0 0 5
0 . 0 2 5
0 . 0 2
0 . 0 1 5
0 . 0 1
0 . 0 0 5
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
g a u s s f i t
0
- 5 0 - 4 0 - 3 0 - 2 0 - 1 0 0 1 0 2 0 3 0 4 0 5 0
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
d a t a
g a u s s fi t
0
- 5 0 -4 0 - 3 0 -2 0 - 1 0 0 1 0 2 0 3 0 4 0 5 0
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
d a t a
g a u s s f i t
0
- 5 0 - 4 0 - 3 0 - 2 0 - 1 0 0 1 0 2 0 3 0 4 0 5 0

G. Peeters A Large Set of Audio Features for Sound Description 2004
6.1.4 Spectral kurtosis (cuidado:AudioSpectrumKurtosis) DS.i_kurto_v
= − μ ⋅ δ ¢
•
•
•
•
•
¢
¡
¡
γ =
σ £
¢
¢
0
- 5 0 - 4 0 - 3 0 - 2 0 -1 0 0 1 0 2 0 3 0 4 0 5 0
23/04/04 14/25
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 . 0 2 5
0 . 0 2
0 . 0 1 5
0 . 0 1
0 . 0 0 5
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
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
d a t a
g a u s s f it
d a ta
g a u s s f i t
- 1 0 -8 - 6 - 4 - 2 0 2 4 6 8 1 0
1 . 2
1
0 . 8
0 . 6
0 . 4
0 . 2
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
d a t a
g a u s s f it
0
- 1 0 -8 - 6 - 4 - 2 0 2 4 6 8 1 0
6.1.5 Spectral slope (cuidado:AudioSpectrumSlope) DS.i_slope_v
$
$ ¤¦¥¨§© £ = ⋅ +
£
=
© ¥ ¡ ¦
−
−

G. Peeters A Large Set of Audio Features for Sound Description 2004
6.1.6 Spectral decrease (cuidado:AudioSpectrumDecrease) DS.i_decr_v
$
$
¢¦ ¤ ¤§¦ © ¦
=
¨
=
£
¨
= £
£
¡
¡
−
−
£
6.1.7 Spectral roll-off (cuidado:AudioSpectrumRollOff) DS.i_rolloff_v
$
$
¤ §©¨
=
¥ ¤
¦
¤
¦
¦
6.2 Temporal variation of spectrum
/ 4# 5 %
. " ; 0 @ - .
8 4 5 8
. " ; 0 @ - .
8
6.2.1 Temporal variation of spectrum: spectral variation (cuidado:AudioSpectrumVariation)
DS.i_var_v
$
=
$
−
−
¤ ¥¦ ¥¤¦
−
⋅
23/04/04 15/25

G. Peeters A Large Set of Audio Features for Sound Description 2004
6.3 Global spectral shape description
6.3.1 Mel Frequency Cepstral Coefficients (MFCC) (cuidado:AudioMFCC) DP.i_MFCC_m
$
$
s(n) FFT MelBand
MFCC
DCT
-, ( - -, $
¦§ ¨¡©£¥£
¨©¥
¢¡¤£¥£ ∂
=
∂
¨©¥ ∂
=
∂
7 Harmonic features
Log
-20
0 1000 2000 3000 4000 5000 6000
1
Frequency
23/04/04 16/25
Log-amplitude
Log-amplitude
Value
-5
-10
-15
0
-1
-2
spectrum
mid-ear spectrum
Mel band spectrum
MFCC spectrum
-3
0 5 10 15 20 25
5
Mel band
MFCC
0
-5
-10
0 2 4 6
MFC coefficient
8 10 12
0 4# 5 -
4 5 ,
,
4 5 ,
7.1.1 Fundamental frequency (mpeg7:AudioFundamentalFrequency) DH.i_f0_v
7.1.2 Noisiness (mpeg7:AudioHarmonicity) DH.i_noisiness_v
=

G. Peeters A Large Set of Audio Features for Sound Description 2004
7.1.3 Inharmonicity (cuidado:AudioInharmonicity) DH.i_inharmo_v
=
¡ ¡
¡ £¢ ¤
−
¡
¤
¡
f0 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
23/04/04 17/25
energy
3 $
( 8
2
7.1.4 Harmonic Spectral Deviation (mpeg7:HarmonicSpectralDeviation)DH.i_devs_v
¤ ¦
§ ©§
= ¥
©
( − )
¦
¤
¦ ¤
¡
¦¨§
¡
¥
Amplitude
0.18
0.16
0.14
0.12
0.1
0.08
0.06
0.04
0.02
Spectral deviation: 0.15374
spectral envelop
harmonics
0
0 2 4 6 8 10
Frequency [harm number]
7 )
8 8

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
$
¡ §£¢
$
=
¥
¥
=
=
¤
¤
¥ ¤
¤ ¤
¤
¤
¤
¤
¡
¡
0
0 5 10
Frequency [harmonic number]
15 20
23/04/04 18/25
Amplitude
0.045
0.04
0.035
0.03
0.025
0.02
0.015
0.01
0.005
Odd/even harmonic energy ratio: 3.2431
: % 8
7.1.6 Tristimulus (cuidado:HarmonicSpectralTristimulus): DH.i_tri*_v
$
¨
¨
§¦
$
= ¥
=
= ¥
=
¤ © ¨
¥
¡
+
¥
¡
¡
¡
+
0.4
0.35
0.3
0.25
0.2
0.15
0.1
0.05
tri1: 0.49442 tri2: 0.45368 tri3: 0.0519
odd harmonic
even harmonic
tristimulus1
tristimulus2
tristimulus3
0
0 5 10 15 20
; 2 (
( 2

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 (cuidado:AudioRelativeSpecificLoudness):
DP.i_specloudnorm_m
=
8.1.3 Sharpness (cuidado:AudioSharpness) DP.i_sharp_v
=
⋅
¦¦¦
=
⋅
¤¡§¦ ¡
= <
=
⋅
⋅
8.1.4 Spread (cuidado:AudioSpread) DP.i_spread_v
=
−
≥
23/04/04 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
•
•
•
•
&
&
=
=
8 & , #
= ⋅
=
* $
£ ¤ ¥
£
∈
©¨
∏
∈
¦¨§¨©¨
©¨¨
¦¨§¨©¨
¢
¡
( ( ∈ ) )
¨
−
¨
∈
¨¨ ¨¨¨
( − μ)
=
23/04/04 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
¨ © © ©
© £
=
+
+
( + − )
©
10.1.4 Temporal modeling an mpeg-7 audio scalable series
mpeg7::scalableseries. weight
scalableseries AudioLoudnessType
mpeg7:scalableseries numOfElements=1 element name
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 2004
©©
¢ ©¢©¨¢
¢¡¤£¦¥¨§©¡¥§©¡©£
¥¨©¢£
£¦¢
©©§©
¡ ¤
©¡¥¨§¥ ¡
©¤
23/04/04 22/25
§¡
§©¢ §©¡
¤©¢ §¥
¤ §¥ ¡
¢ ¢©©¢
¢ ¢©© ¢
¢ ¢©©¢ ¢ ¢¢
¢ ¢©©
' = # $ ! * " #

G. Peeters A Large Set of Audio Features for Sound Description 2004
11 List of all descriptors
LLD List
frame
based
number of
features acronym xml tag
Temporal Features
Global Temporal Features
Log Attack Time n 1 DTg_lat mpeg7:LogAttackTime
Temporal Increase n 1 DTg_incr cuidado:TemporalIncrease
Temporal Decrease n 1 DTg_decr cuidado:TemporalDecrease
Temporal Centroid n 1 DTg_tc mpeg7:TemporalCentroid
Effective Duration n 1 DTg_ed cuidado::TemporalEffectiveDuration
Instantaneous Temporal Features
Signal Auto-correlation function y 12 DTi_xcorr_m cuidado:AudioXcorr
Zero-corssing rate y 1 DTi_zcr cuidado:AudioZcr
Energy Features
Total energy y 1 DEi_tot_v mpeg7:AudioPower
Total energy Modulation (frequency, amplitude) n 2 DTg_mod_fr, DTg_mod_am ScalableSeriesType element name="Modulation"
Total harmonic energy y 1 DEi_harmo_v cuidado:AudioHarmonicPower
Total noise energy y 1 DEi_noise_v cuidado:AudioNoisePower
Spectral Features
Spectral Shape
Spectral centroid y 6 DSi_sc_m mpeg7:AudioSpectrumCentroid (mpeg7:SpectralCentroid)
Spectral spread y 6 DSi_ss_m mpeg7:AudioSpectrumSpread
Spectral skewness y 6 Dsi_skew_m cuidado:AudioSpectrumSkewness
Spectral kurtosis y 6 Dsi_kurto_v cuidado:AudioSpectrumKurtosis
Spectral slope y 6 Dsi_slope_v cuidado:AudioSpectrumSlope
Spectral decrease y 1 Dsi_decs_c cuidado:AudioSpectrumDecrease
Spectral rolloff y 1 Dsi_rolloff_v cuidado:AudioSpectrumRollOff
Spectral variation y 3 Dsi_variation_v cuidado:AudioSpectrumVariation
Global spectral shape description
MFCC y 12 DPi_mfcc_m cuidado:AudioMFCC
Delta MFCC y (post) 12 DPi_Dmfcc_m
Delta Delta MFCC y (post) 12 DPi_DDmfcc_m
Harmonic Features
Fundamental frequency y 1 DHi_f0_v mpeg7:AudioFundamentalFrequency
Fundamental fr. Modulation (frequency, amplitude) n 2 F0 Mod AM, FR ScalableSeriesType element name="Modulation"
Noisiness y 1 DHi_noisiness_v mpeg7:AudioHarmonicity
Inharmonicity y 1 DHi_inharmo_v cuidado:AudioInharmonicity
Harmonic Spectral Deviation y 3 DHi_devs_v mpeg7:HarmonicSpectralDeviation
Odd to Even Harmonic Ratio y 3 Dhi_oeratio_v cuidado:HarmonicSpectralOERatio
Harmonic Tristimulus y 9 Dhi_tri_v cuidado:HarmonicSpectralTristimulus
Harmonic Spectral Shape
HarmonicSpectral centroid y 6 DHi_sc_m mpeg7:HarmonicSpectralCentroid
HarmonicSpectral spread y 6 DHi_ss_m mpeg7:HarmonicSpectralSpread
HarmonicSpectral skewness y 6 DHi_skew_m cuidado:HarmonicSpectralSkewness
HarmonicSpectral kurtosis y 6 DHi_kurto_v cuidado:HarmonicSpectralKurtosis
HarmonicSpectral slope y 6 DHi_slope_v cuidado:HarmonicSpectralSlope
HarmonicSpectral decrease y 1 DHi_decs_c cuidado:HarmonicSpectralDecrease
HarmonicSpectral rolloff y 1 DHi_rolloff_v cuidado:HarmonicSpectralRollOff
HarmonicSpectral variation y 3 DHi_variation_v mpeg7:HarmonicSpectralVariation
Perceptual Features
Loudness y 1 DPi_loud_v AudioLoudness
RelaitveSpecific Loudness y 24 DPi_specloud_m cuidado:AudioRelativeSpecificLoudness
Sharpness y 1 DPi_sharp_v cuidado:AudioSharpness
Spread y 1 DPi_spread_v cuidado:AudioSpread
Perceptual Spectral Envelope Shape
Perceptual Spectral centroid y 6 DPi_sc_m cuidado:AudioFilterbankCentroid
Perceptual Spectral spread y 6 DPi_ss_m cuidado:AudioFilterbankSpread
Perceptual Spectral skewness y 6 DPi_skew_m cuidado:AudioFilterbandSkewness
Perceptual Spectral kurtosis y 6 DPi_kurto_v cuidado:AudioFilterbankKurtosis
Perceptual Spectral Slope y 6 DPi_slope_v cuidado:AudioFilterbankSlope
Perceptual Spectral Decrease y 1 DPi_decs_c cuidado:AudioFilterbankDecrease
Perceptual Spectral Rolloff y 1 DPi_rolloff_v cuidado:AudioFilterbankRolloff
Perceptual Spectral Variation y 3 DPi_variation_v cuidado:AudioFilterbankVariation
Odd to Even Band Ratio y 3 DP_ioeratio_v cuidado:AudioFilterbankOERatio
Band Spectral Deviation y 3 DPi_devs_v cuidado:AudioFilterbankDeviation
Band Tristimulus y 9 DPi_tri_v cuidado:AudioFilterbankTristimulus
Various features
Spectral flatness y 4 DPi_sfm_m mpeg7:AudioSpectrumFlatness
Spectral crest y 4 DPi_scm_m cuidado:AudioSpectrumCrest
Total Number of Features 166
23/04/04 23/25

G. Peeters A Large Set of Audio Features for Sound Description 2004
12 Acknowledgement
13 References
£ ¤£¦¥¨§©¥¨ ¢£ ¨ ¡ £ ¦¡ £ ¥ £ £ ¦ ¦£ £¤£¨¨¡¦ £ £ ¤£¡¤¥¥ ¡ £ ¦¥¨£¥ ¢£ ¡¨ £ ¥ £ £ £
¨£¢¡
¤£¦ ¤¦¥
£ £¦¥¢§©¢¥¢§£ ¡¨ ¡ £ ¥ £ £ £¥ £ ¨ ¡ £ ¦ £ ¡¤¨ £ ¨¡¤£ £ £ £ ¡ ¥ ¤¡ ¨¤£¡¤¥¥ ¡ £
¨£¢¡
¥¡¨¡¤¥ £
= 0 §¢©¢©¢¨©¥
¡ ¡¥ ¥ ¤¨ ¡¤¥¢§©¢¢¥¨ £ £¥ ¤¨ ¢¥©£¨ ¡£ £ £ ¦¥¡¤ £ ¦¡ £ ¤¨£ £¡¥
§ ££ ¡ £ © ¥
¢
£ ¥§©¢¢¥ ¤ £ £ ¡ ¥¨¡ £ ¤¡ ¥£ £ ¡ £ ¡¡¤ £ £ £ £¨¡¤¥ ¨ © £ ¡¤¨ ¤¨ ¢ ¥
£ ¥ £ ¥ £¡§©¢¢¥ £ ¡ £ ¥¨¡¡ £ ¤¡¤ £ £ £ ¨¤ £ £ ¦ ¦ ¦¡ ¥££ ¥¤¨ £ £
¤¥ £ £ ¦¥ ¢ § ££ ¡ £ ¦¥
§£ £ ¡ £ ¡ ¡¤¨ ¡ £¥ £ £¥ ¡ £¨¡ £ ¨¥ ¦£ ¡ ¢¤¢
££¡£¦¥§©¢¢¥
£ £¥ ¢ ¥
¨£¤¡¤¡
£¡ £ ¡ £ ¥§¢¨©¥ ¥¡ ¦ ¨¡ £¦¡¥£ ¤¨ ¤¨¤¡£ £ ¨ ¦£ ¥
¥£¥¥ £¦¥ ¥¢ ¡¤ ¥¢§©¢¥¢§¤¨¤¤¨ £ £ ¨¡ ¦¡¤¢£ £ ¤£ ¡¡¤¥ ¢¤¢ £ ¡ £¦ ¡ ¡¤
¢¨
£ ¥ £ £ ¥£ ¥ ¦¡¥ ££¨ £ ¡ ¥ ¡
/ ¢¢¢¨¥
¤¨ £ ¥ £ ¥ §©¥ ©¨¥ ¢ £ ¨ ¡ £ ¡ £ ¥ £ £ ¤¡¨ £ §¨¡¤£ £ £ £ ¨£¥¥©¢¦¥¡¤¡¤¥ ¥£¥
¡¤¨ ¥ £¥
¤¨ ¦ ¥ ¥ ¦¥ ¡¤¥§©¢¨¥ ¤ £ £ £¥ ¡ £ £ £ £ ¤¡ £¦¡¥£¨ ¡¤¨¡ ¦ ¤¨ ¡
£ £ ¦ ¨¡¤¦¥©¥¡¤¡¤¥ ¥£¥£¥ ¡¤¨ ©¥ ¢£ ¥§£ £ ¥£ £ ¤£¦ ¤¡¡ ¥ £ £ £ © ¥
¦ ¥¢§©¢¢¥¢ £¦¡¥£¨¢¦¥¡¨¡ ¤ £ £ £¥ ¥¦¨¡¤¥££ £ ¡¤ £ ¤¨ £ £ ¡¤¥ £¥ £
££¨¡ ¡¤¨¡¤
¥¢£¥ ¢¢¢¥
£
/ 0
§¢¢¥ ¢£ ¥£¨ £ £ ¡¥ £ ££ ¡ §£ £ ¡ £ ¡¤¨ £ £ ¢£ ¡¨¥¤¡¢ ¤¨ £¥ ¢ ¢ §£ §
§¥ ¢ ¢ § ¢ ©¢ ¢¥
©
¥§¢¥ £ § ¥ £ £ £¥ ¨¤¦¡ ¢£ ¨ ¡ £ ¤¡¤ £ ¡¤¨¤¨¥ § ¥ ¡¨¡ ¥
¡¤¡¤¡¤¨
¡¨ £ ¤ £ £ ¥ ©©¦¥ §£ £ ¡ £ £ £ ¢¡¡¢£¨© ¥
¢¤£
¥ ¥ ¡¤ ¥§¢¢¥ ¢£ ¨¤ ¡ £ £ £ ¡¤¨ £ £ £ ¦¥¡ ¤£ £ ¡£¥ ¥ ¢ § § ¡¤¨ ¦£
¡¤¡¤¡¤¨
£ ¥ ¢¡¤¨
¥ £ ¥ £¦¡¤§¢¢¥ £ ¤ ¡¡¤ £ £ ¦¡¤¨ £¨ ¥£¨£ £ §¨ ¥ £ £ ¥ ¢ § § ¢£¡£¨
¡¤¡¤¡¤¨
£ ¥ ¡¢¡¦¡
¥ £ ¥£¦¡¤§¢¢¥ ¡¤¨¥ ¦ £ ¨¡¤¡¡ ¥ ¢£ ¡¨ £ ¦ ¦ ¤ £ £ ¥£¨¢¦¥¡ §¦ ¥ £ £ £¥
¡¤¡¤¡¤¨
¨¦¡ ¢£ ¨¤ ¡ £ ¡¥ £ £ ¦£ £ ©¥
£ ¡¤¨¤¥ £ ¥£ £ §©¢¢¥ £ ¡ £ £¥ ¡¡¥¨¡£ £ £ £ ¥¢¡¡ £¨¨¡ £ ¤¡ ¥
¥¡ ¨¡¤¨¤¤¥ £ ¥ £ ¡¤§©¢¢¥ §£ £ ¨¤ £ £ £ ¡ £ £ £¥¨£¦ ¥¡¨¡ ¡¤¡¥ ¨ £ £¨¥
§ £ ¥ ¢¡¨ £ ¥
¢
23/04/04 24/25

G. Peeters A Large Set of Audio Features for Sound Description 2004
£ £¦¥ £ £ §©¢¨¥£ £ ¨ £ ¥©£¨ £ £ ¡ £ £ ¥ ¡ ¡ ¨ ¦¡¦¥ ¨¤¡£ £
¡¤¨¨¥
£ ¥ §
£¤¥ ¤¥ ¡ ¥§©¢¥ § ¥ £ £ ¡¨¥ £ ¨¡¤¨ ¡ £¥ £¥ §§ £ £¦£ £¥ ¡
£ ¥ ¥ ¨¥¥ ££ £ §§ ¨¡¤ §£ ¢¢¢¥
$
¥£¦¡¤¥¤¥§¢©¥¨£¡¤ ¨ £ ¡¤¡¤¨ ¡¨¡ £ ¡ ¥¤ ¤¨ ¢ ¨ ¥
¡ ¡ ¡¤¨¤¥§©¢¢¥¨£¡¦¨¡ ¥£¨¤ £ £ £ ¡ £¥¡ £¨ ¨ ¡ £ £ ¦¦¥ £ ¥
¡ ¡ ¡¤¨¤¥§©¢¢¥¥£ £ ¤¥ ¡¨ ¦£ ¨ £ ¡¨¢¡¨¥
¡ ¡¤¨¤¥ £ ¤¥¡¤¨¤¥¤¨§©¨¢¥ £ ¡ ¨¡¤ £ £ ¥£¨¤¨ £ ¨¤¡ £ ¨ £ ¢¡ ¦¥¢¥ £ ¤ £ £ £¥
¡
£ ¥ £
¥¨¡¡
3 : ©¢¢©¢¥
23/04/04 25/25