7- Speech Quality Assessment

7-Speech Quality Assessment

Quality Levels

Subjective TestsObjective Tests

IntelligibilityNaturalness

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)

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

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

First Class (Cont’d)Subjective Intelligibility tests

DRT is very applicable and credibleIn this test user can hear the speech only once

100%

Tests

IncorrectCorrect

NNN

DRT

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

Mean Opinion Score (MOS)

Scores for MOS are like this

Score Speech Quality1

2

3

4

5

Not Acceptable

Weak

Medium

Good

Excellent

Diagnostic Acceptability Measure (DAM)

This test is very complexIn this test there is 19 different parameters for score. These parameters divide into 3 main groups:– Signal Quality– Background Quality– Total Quality

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

Objective Tests (Cont’d)

Articulation Index (AI)

Signal to Noise Ratio (SNR)– Global (Classic) SNR– Segmental SNR– Frequency Weighted Segmental SNR

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

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

Articulation index (Cont’d)

In AI SNR measured isolated in each band

20

1 30)30,(

201

j

SNRMinAI

Signal To Noise Ratio(SNR)

)()()( ˆ nnn ss

n

nnn

n ssE 2)()(

2)( ]ˆ[

n

ns sE 2)(

nnn

nn

sglobal

ss

s

EE

SNR2

)()(

2)(

)(

]ˆ[log10log10

Segmental SNR

1

0

1

2)()(

1

2)(

)( ]]ˆ[

[log101 M

jm

Nmnnn

m

Nmnn

seg j

j

j

j

ss

s

MSNR

j’th Frame SNR

M : Number of frames

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

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

M w

Other Formulas:

1,

( ) 10 ,0 1 ,

,1

( )1 1 10log( )

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

M w

Itakura Measure

)(H

)(S

)(H Is the envelope spectrum

2|)(|)()}({)( XSRFS

Use from All-Pole (AR) Model

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

Itakura Measure (Cont’d)

M

lssss mlgmlg

Mmgmgd

1

2ˆˆ )],(),([1))(),((

m :Index of frame

l : Index of coefficients

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

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