7-speech quality assessment quality levels subjective tests objective tests...
TRANSCRIPT
![Page 1: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/1.jpg)
7-Speech Quality Assessment
Quality Levels
Subjective Tests
Objective Tests
Intelligibility
Naturalness
![Page 2: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/2.jpg)
Quality Levels
Synthetic Quality (Under 4.8 kbps)
Communication Quality (4.8 to 13 kbps)
Toll Quality (13 to 64 kbps)
Broadcast Quality (Upper than 64 kbps)
![Page 3: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/3.jpg)
Test Types
Intelligibility Naturalness
Subjective DRT, MRT MOS, DAM
Objective None.Future ASR systems
AI, Global SNR, Seg. SNR, FW-Seg. SNR, Itakura Measure,WSSM
![Page 4: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/4.jpg)
First ClassSubjective Intelligibility Tests
Diagnostic Rhyme Test (DRT)– Selecting between two CVC by different first C– First C should have specific properties– Ex. hop - fop And than - dan
Modified Rhyme Test (MRT)– Selecting between CVC’s by different first C– Ex. Cat, bat, rat, mat, fat, sat
![Page 5: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/5.jpg)
First Class (Cont’d)Subjective Intelligibility tests
DRT is very applicable and credible
In this test user can hear the speech only once
100%
Tests
IncorrectCorrect
N
NNDRT
![Page 6: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/6.jpg)
Second ClassSubjective Naturalness tests
Mean Opinion Score (MOS)– MOS is very applicable and credible– In this test user can hear the speech a lot
Diagnostic Acceptability Measure (DAM)– This test is very complex
![Page 7: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/7.jpg)
Mean Opinion Score (MOS)
Scores for MOS are like this
Score Speech Quality1
2
3
4
5
Not Acceptable
Weak
Medium
Good
Excellent
![Page 8: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/8.jpg)
Diagnostic Acceptability Measure (DAM)
This test is very complex
In this test there is 19 different parameters for score. These parameters divide into 3 main groups:– Signal Quality– Background Quality– Total Quality
![Page 9: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/9.jpg)
Objective Tests
These tests can not be used for intelligibility. Because system couldn’t recognize speech intelligibility
Objective tests can only be used for speech Naturalness
![Page 10: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/10.jpg)
Objective Tests (Cont’d)
Articulation Index (AI)
Signal to Noise Ratio (SNR)– Global (Classic) SNR– Segmental SNR– Frequency Weighted Segmental SNR
![Page 11: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/11.jpg)
Articulation Index (AI)
AI assumes that different frequency bands distortion are independent, and measure signal quality in different bands.
In each band determines percentage of perceptible signal by listener
. . . . . . . . . 20 BandsHZ
200 6100
![Page 12: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/12.jpg)
Articulation index (Cont’d)
Perceptible by user signal :– 1- Upper than human hearing threshold– 2- Under than human pain threshold– 3- Upper than Masking Noise level
– In each case one of the states 1 or 3 is prevail
![Page 13: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/13.jpg)
Articulation index (Cont’d)
In AI SNR measured isolated in each band
20
1 30
)30,(
20
1
j
SNRMinAI
![Page 14: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/14.jpg)
Signal To Noise Ratio(SNR)
)()()( ˆ nnn ss
n
nnn
n ssE 2)()(
2)( ]ˆ[
n
ns sE 2)(
nnn
nn
sglobal
ss
s
E
ESNR
2)()(
2)(
)(
]ˆ[log10log10
![Page 15: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/15.jpg)
Segmental SNR
1
0
1
2)()(
1
2)(
)( ]
]ˆ[
[log101 M
jm
Nmnnn
m
Nmnn
segj
j
j
j
ss
s
MSNR
j’th Frame SNR
M : Number of frames
![Page 16: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/16.jpg)
Frequency Weighted Segmental SNR
1
0
1,
1,,,
)( ]])()([
log[101 M
jK
kkj
K
kjkjkskj
segfw
W
mEmEW
MSNR
K : Number of frequency bands
M : Number of frames
![Page 17: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/17.jpg)
Deller Formula
, 10 , ,11
( ) 100
,1
10log [ ( ) ( )]1
10log [ ]
K
j k s k j k jMk
fw seg Kj
j kk
w E m E mSNR
Mw
![Page 18: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/18.jpg)
Other Formulas:
1,
( ) 10 ,0 1 ,
,1
( )1 110log
( )
M Ks k j
fw seg j kKj k k j
j kk
E mSNR w
M E mw
, 10 , ,11
( )0
,1
10log [ ( ) ( )]1
K
j k s k j k jMk
fw seg Kj
j kk
w E m E mSNR
Mw
![Page 19: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/19.jpg)
Itakura Measure
)(H
)(S
)(H Is the envelope spectrum
2|)(|)()}({)( XSRFS
Use from All-Pole (AR) Model
![Page 20: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/20.jpg)
Itakura Measure (Cont’d)
p
i
jiea
H
1
1
1)(
This is based on the spectrum difference between main signal and assessment signal
ia
iRiK
Autoregressive Coefficients
Reflection Coefficients
Autocorrelation Coefficients
![Page 21: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/21.jpg)
Itakura Measure (Cont’d)
M
lssss mlgmlg
Mmgmgd
1
2ˆˆ )],(),([
1))(),((
m :Index of frame
l : Index of coefficients
![Page 22: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/22.jpg)
Itakura Measure (Cont’d)
1
1',,
1ˆ',,
ˆ
])]',(),([
[
))'(),((~
M
lmml
M
lssmml
sslp
W
mlmlW
mmd
),( mls Is the l’th parameter of the frame that conduces m’th sample
![Page 23: 7-Speech Quality Assessment Quality Levels Subjective Tests Objective Tests IntelligibilityNaturalness](https://reader038.vdocuments.site/reader038/viewer/2022110321/56649f505503460f94c7246c/html5/thumbnails/23.jpg)
Weighted Spectral Slope Measure(WSSM)
|),(||),1(||),(| mksmksmks |),(ˆ||),1(ˆ||),(ˆ| mksmksmks
236
1, ]|),(ˆ||),(|[
|)),(ˆ||,),((|
k
mk
WSSM
mksmksWK
msmsd
),( mks Is STFT of k’th band of the frame that conduces m’th sample
dB.in are|),(||),1(| mksandmks