SIST-V ETSI/EG 202 396-3 V1.3.0:2011

ETSI Guide
Speech and multimedia Transmission Quality (STQ);
Speech Quality performance
in the presence of background noise
Part 3: Background noise transmission -
Objective test methods
2 ETSI EG 202 396-3 V1.3.1 (2011-02)

Reference
REG/STQ-00167
Keywords
noise, QoS, quality, speech
ETSI
650 Route des Lucioles
F-06921 Sophia Antipolis Cedex - FRANCE

Tel.: +33 4 92 94 42 00  Fax: +33 4 93 65 47 16

Siret N° 348 623 562 00017 - NAF 742 C
Association à but non lucratif enregistrée à la
Sous-Préfecture de Grasse (06) N° 7803/88

Important notice
Individual copies of the present document can be downloaded from:
http://www.etsi.org
The present document may be made available in more than one electronic version or in print. In any case of existing or
perceived difference in contents between such versions, the reference version is the Portable Document Format (PDF).
In case of dispute, the reference shall be the printing on ETSI printers of the PDF version kept on a specific network drive
within ETSI Secretariat.
Users of the present document should be aware that the document may be subject to revision or change of status.
Information on the current status of this and other ETSI documents is available at
http://portal.etsi.org/tb/status/status.asp
If you find errors in the present document, please send your comment to one of the following services:
http://portal.etsi.org/chaircor/ETSI_support.asp
Copyright Notification
No part may be reproduced except as authorized by written permission.
The copyright and the foregoing restriction extend to reproduction in all media.

© European Telecommunications Standards Institute 2011.
All rights reserved.
TM TM TM TM
DECT , PLUGTESTS , UMTS , TIPHON , the TIPHON logo and the ETSI logo are Trade Marks of ETSI registered
for the benefit of its Members.
TM
3GPP is a Trade Mark of ETSI registered for the benefit of its Members and of the 3GPP Organizational Partners.
LTE™ is a Trade Mark of ETSI currently being registered
for the benefit of its Members and of the 3GPP Organizational Partners.
GSM® and the GSM logo are Trade Marks registered and owned by the GSM Association.
ETSI
3 ETSI EG 202 396-3 V1.3.1 (2011-02)
Contents
Intellectual Property Rights . 5
Foreword . 5
1 Scope . 6
2 References . 6
2.1 Normative references . 6
2.2 Informative references . 7
3 Abbreviations . 8
4 Speech signals to be used . 9
5 Selection of the data within the scope of the wideband objective model: Experts evaluation . 9
5.1 Selection process . 9
5.2 Results . 10
5.3 French database . 10
5.4 Czech database . 10
5.5 General differences between the databases . 12
6 Description of the wideband objective test method . 16
6.1 Introduction . 16
6.2 Speech sample preparation and nomenclature . 17
6.2.1 Speech sample preparation . 17
6.2.2 Nomenclature . 19
6.3 Principles of Relative Approach and Δ Relative Approach . 20
6.4 Objective N-MOS. 23
6.4.1 Introduction. 23
6.4.2 Description of N-MOS algorithm . 24
6.4.3 Comparing subjective and objective N-MOS results . 27
6.5 Objective S-MOS . 28
6.5.1 Introduction. 28
6.5.2 Description of S-MOS Algorithm . 28
6.5.3 Comparing Subjective and Objective S-MOS Results . 32
6.6 Objective G-MOS. 32
6.6.1 Description of G-MOS Algorithm . 32
6.6.2 Comparing subjective and objective G-MOS results . 33
6.7 Comparison of the objective method results for Czech and French samples . 34
6.8 Language Dependent Robustness of G-MOS . 38
7 Validation of the Wideband Objective Test Method . 40
7.1 Introduction . 40
7.2 All conditions results analysis . 42
7.2.1 Comparing subjective and objective N-MOS results . 42
7.2.2 Comparing subjective and objective S-MOS results . 43
7.2.3 Comparing Subjective and Objective G-MOS Results . 43
7.3 French Conditions Results Analysed . 44
7.3.1 Comparing Subjective and Objective N-MOS Results . 44
7.3.2 Comparing Subjective and Objective S-MOS Results . 45
7.3.3 Comparing subjective and objective G-MOS results . 45
7.4 Czech conditions results analysis . 46
7.4.1 Comparing subjective and objective N-MOS results . 46
7.4.2 Comparing subjective and objective S-MOS results . 47
7.4.3 Comparing Subjective and Objective G-MOS Results . 47
8 Objective Model for Narrowband Applications . 48
8.1 File pre-processing . 48
8.2 Adaptation of the Calculations . 49
ETSI
4 ETSI EG 202 396-3 V1.3.1 (2011-02)
Annex A: Detailed post evaluation of listening test results . 51
Annex B: Results of PESQ and TOSQA2001 - Analysis of EG 202 396-2 database . 56
Annex C: Comparison of objective MOS versus auditory MOS for the All Data of Training
Period . 63
Annex D: Comparison of objective MOS versus auditory MOS for the Data not used during
the Training Period . 65
Annex E: Regression Coefficients for Czech data . 67
Annex F: Detailed STF 294 subjective and objective validation test results . 68
Annex G: Void . 72
Annex H: Extension of the EG 202 396-3 Speech Quality Test Method to Narrowband:
Adaptation, Training and Validation . 73
Annex I: Validation results of the modified EG 202 396-3 objective speech quality model for
narrowband data . 77
I.1 Introduction . 77
I.2 Description of the Databases . 77
I.3 Collection of the subjective scores . 78
I.4 Differences: HEAD acoustics training database vs. France Telecom validation databases . 80
I.5 Results . 81
I.6 Unmapped Results . 81
I.7 Mapped Results . 84
I.7.1 Use of mapping functions . 84
I.8 Conclusions . 90
History . 92

ETSI
5 ETSI EG 202 396-3 V1.3.1 (2011-02)
Intellectual Property Rights
IPRs essential or potentially essential to the present document may have been declared to ETSI. The information
pertaining to these essential IPRs, if any, is publicly available for ETSI members and non-members, and can be found
in ETSI SR 000 314: "Intellectual Property Rights (IPRs); Essential, or potentially Essential, IPRs notified to ETSI in
respect of ETSI standards", which is available from the ETSI Secretariat. Latest updates are available on the ETSI Web
server (http://webapp.etsi.org/IPR/home.asp).
Pursuant to the ETSI IPR Policy, no investigation, including IPR searches, has been carried out by ETSI. No guarantee
can be given as to the existence of other IPRs not referenced in ETSI SR 000 314 (or the updates on the ETSI Web
server) which are, or may be, or may become, essential to the present document.
Foreword
This ETSI Guide (EG) has been produced by ETSI Technical Committee Speech and multimedia Transmission Quality
(STQ).
The present document is a deliverable of ETSI Specialized Task Force (STF) 294 entitled: "Improving the quality of
eEurope wideband speech applications by developing a performance testing and evaluation methodology for
background noise transmission".
The present document is part 3 of a multi-part deliverable covering Speech and multimedia Transmission Quality
(STQ); speech quality performance in the presence of background noise, as identified below:
Part 1: "Background noise simulation technique and background noise database";
Part 2: "Background noise transmission - Network simulation - Subjective test database and results";
Part 3: "Background noise transmission - Objective test methods".
ETSI
6 ETSI EG 202 396-3 V1.3.1 (2011-02)
1 Scope
The present document aims to identify and define testing methodologies which can be used to objectively evaluate the
performance of narrowband and wideband terminals and systems for speech communication in the presence of
background noise.
Background noise is a problem in mostly all situations and conditions and need to be taken into account in both,
terminals and networks. The present document provides information about the testing methods applicable to objectively
evaluate the speech quality in the presence of background noise. The present document includes:
• The description of the experts post evaluation process chosen to select the subjective test data being within the
scope of the objective methods.
• The results of the performance evaluation of the currently existing methods described in ITU-T
Recommendation P.862 [i.16], [i.17] and in TOSQA2001 [i.19] which is chosen for the evaluation of terminals
in the framework of ETSI VoIP speech quality test events [i.8], [i.9], [i.10] and [i.11].
• The method which is applicable to objectively determine the different parameters influencing the speech
quality in the presence of background noise taking into account:
- the speech quality;
- the background noise transmission quality;
- the overall quality.
• The document is to be used in conjunction with:
- EG 202 396-1 [i.1] which describes a recording and reproduction setup for realistic simulation of
background noise scenarios in lab-type environments for the performance evaluation of terminals and
communication systems.
- EG 202 396-2 [i.2] which describes the simulation of network impairments and how to simulate realistic
transmission network scenarios and which contains the methodology and results of the subjective scoring
for the data forming the basis of the present document.
- French speech sentences as defined in ITU-T Recommendation P.501 [i.13] for wideband and English
speech sentences as defined in ITU-T Recommendation P.501 [i.13] for narrowband.
2 References
References are either specific (identified by date of publication and/or edition number or version number) or
non-specific. For specific references, only the cited version applies. For non-specific references, the latest version of the
reference document (including any amendments) applies.
Referenced documents which are not found to be publicly available in the expected location might be found at
http://docbox.etsi.org/Reference.
NOTE: While any hyperlinks included in this clause were valid at the time of publication ETSI cannot guarantee
their long term validity.
2.1 Normative references
The following referenced documents are necessary for the application of the present document.
Not applicable.
ETSI
7 ETSI EG 202 396-3 V1.3.1 (2011-02)
2.2 Informative references
The following referenced documents are not necessary for the application of the present document but they assist the
user with regard to a particular subject area.
[i.1] ETSI EG 202 396-1: "Speech and multimedia Transmission Quality (STQ); Speech quality
performance in the presence of background noise; Part 1: Background noise simulation technique
and background noise database".
[i.2] ETSI EG 202 396-2: "Speech Processing, Transmission and Quality Aspects (STQ); Speech
Quality performance in the presence of background noise; Part 2: Background Noise Transmission
- Network Simulation - Subjective Test Database and Results".
[i.3] ITU-T Recommendation P.835: "Subjective test methodology for evaluating speech
communication systems that include noise suppression algorithm".
[i.4] ITU-T Recommendation P.800: "Methods for subjective determination of transmission quality".
[i.5] ITU-T Recommendation P.831: "Subjective performance evaluation of network echo cancellers".
[i.6] Genuit, K.: "Objective Evaluation of Acoustic Quality Based on a Relative Approach", InterNoise
'96, Liverpool, UK.
[i.7] ITU-T Recommendation SG 12 Contribution 34: "Evaluation of the quality of background noise
transmission using the "Relative Approach"".
[i.8] ETSI 2nd Speech Quality Test Event: "Anonymized Test Report", ETSI Plugtests, HEAD
acoustics, T-Systems Nova.
NOTE: Available at: http://www.etsi.org/WebSite/OurServices/Plugtests/History.aspx.
Also available as ETSI TR 102 648-3.
[i.9] ETSI 3rd Speech Quality Test Event: "Anonymized Test Report "IP Gateways"".
NOTE: Available at: http://www.etsi.org/WebSite/OurServices/Plugtests/History.aspx.
[i.10] ETSI 3rd Speech Quality Test Event: "Anonymized Test Report "IP Phones"".
[i.11] ETSI 4th Speech Quality Test Event: "Anonymized Test Report "IP Gateways and IP Phones"".
NOTE: Available at: http://www.etsi.org/WebSite/OurServices/Plugtests/History.aspx.
[i.12] F. Kettler, H.W. Gierlich, F. Rosenberger: "Application of the Relative Approach to Optimize
Packet Loss Concealment Implementations", DAGA, March 2003, Aachen, Germany.
[i.13] ITU-T Recommendation P.501: "Test Signals for Use in Telephonometry".
[i.14] R. Sottek, K. Genuit: "Models of Signal Processing in human hearing", International Journal of
Electronics and Communications (AEÜ) vol. 59, 2005, p. 157-165.
NOTE: Available at: http://www.elsevier.de/aeue.
[i.15] SAE International - Document 2005-01-2513: "Tools and Methods for Product Sound Design of
Vehicles" R. Sottek, W. Krebber, G. Stanley.
[i.16] ITU-T Recommendation P.862: "Perceptual evaluation of speech quality (PESQ): An objective
method for end-to-end speech quality assessment of narrowband telephone networks and speech
codecs".
[i.17] ITU-T Recommendation P.862.1: "Mapping function for transforming P.862 raw result scores to
MOS-LQO".
[i.18] ITU-T Recommendation P.862.2: "Wideband extension to Recommendation P.862 for the
assessment of wideband telephone networks and speech codecs".
ETSI
8 ETSI EG 202 396-3 V1.3.1 (2011-02)
[i.19] ITU-T Recommendation SG 12 Contribution 19: "Results of objective speech quality assessment
of wideband speech using the Advanced TOSQA2001".
[i.20] ITU-T Recommendation G.722: "7 kHz audio-coding within 64 kbit/s".
[i.21] ITU-T Recommendation G.722.2: "Wideband coding of speech at around 16 kbit/s using Adaptive
Multi-Rate Wideband (AMR-WB)".
[i.22] ITU-T Recommendation P.56: "Objective measurement of active speech level".
[i.23] ITU-T Recommendation P.57: "Artificial ears".
[i.24] M. Spiegel: "Theory and problems of statistics", McGraw Hill, 1998.
[i.25] R.A. Fisher: "Statistical methods and scientific inference", Oliver and Boyd, 1956.
[i.26] M. Kendall: "Rank correlation methods", Charles Griffin & Company Limited, 1948.
[i.27] Sottek, R.: "Modelle zur Signalverarbeitung im menschlichen Gehör, PHD thesis RWTH Aachen,
1993".
[i.28] ITU-T Recommendation P.830: "Subjective performance assessment of telephone-band and
wideband digital codecs".
[i.29] ITU-T contribution COM 12-117, Study Period 1997-2000: "Report of the question 13/12
rapporteur's meeting (Solothurn, Germany, 6-10 March 2000)".
[i.30] ANSI S1.1-1986 (ASA 65-1986), "Specifications for Octave-Band and Fractional-Octave-Band
Analog and Digital Filters", 1993.
3 Abbreviations
For the purposes of the present document, the following abbreviations apply:
ACR Absolute Comparison Rating
AMR Adaptive MultiRate
ASL Active Speech Level
NOTE: According to ITU-T Recommendation P.56 [i.22].
BGN BackGround Noise
CDF Cumulative Density Function
DB Data Base
dB SPL Sound Pressure Level re 20 µPa in dB
G-MOS Global MOS
NOTE: MOS related to the overall sample.
HP HighPass
IP Internet Protocol
IRS Intermediate Reference System
ITU International Telecommunication Union
ITU-T Telecom Standardization Body of ITU
MOS Mean Opinion Score
MOS-LQSN Mean Opinion Score - Listening Quality Subjective Noise
MRP Mouth Reference Point
NI Network I conditions
NII Network II conditions
NIII Network III conditions
NB NarrowBand
N-MOS Noise MOS
NOTE: MOS related to the noise transmission only.
ETSI
9 ETSI EG 202 396-3 V1.3.1 (2011-02)
NR Noise Reduction
NR (filter) Noise Reduction (filter)
PESQ Perceptual Evaluation of Speech Quality
PLC Packet Loss Concealment
RCV ReCeiVe
RMSE Random Mean Square Error
S-MOS Speech MOS
NOTE: MOS related to the speech signal only.
SNR Signal to Noise Ratio
STF Specialized Task Force
TMOS TOSQA Mean Opinion Score
TOR Terms Of Reference
TOSQA Telecommunication Objective Speech Quality Assessment
VAD Voice Activity Detection
VoIP Voice over IP
WB WideBand
4 Speech signals to be used
As with any objective model, the prediction of speech quality depends on the conditions under which the model was
tested and validated (see clauses 6.1 and 8). This dependency also applies to the speech material used in conjunction
with the objective model.
The wideband version of the model uses French speech sentences. The near end speech signal (clean speech signal)
consists of 8 sentences of speech (2 male and 2 female talkers, 2 sentences each). Appropriate speech samples can be
taken from ITU-T Recommendation P.501 [i.13].
The narrowband version of the model uses English speech sentences. The near end speech signal (clean speech signal)
consists of 8 sentences of speech (2 male and 2 female talkers, 2 sentences each). Appropriate speech samples can be
taken from ITU-T Recommendation P.501 [i.13].
5 Selection of the data within the scope of the
wideband objective model: Experts evaluation
5.1 Selection process
The aim of the selection process was to identify those data in the databases described in EG 202 396-2 [i.2] which are
consistent with the scope of the objective models to be studied within the present document.
The experts were selected on the based on the definition found in e.g. ITU-T Recommendation P.831 [i.5]: experts are
experienced in subjective testing. Experts are able to describe an auditory event in detail and are able to separate
different events based on specific impairments. They are able to describe their subjective impressions in detail. They
have a background in technical implementations of noise reduction systems and transmission impairments and do have
detailed knowledge of the influence of particular implementations on subjective quality.
Their task was to select the relevant conditions within the scope of the model to be developed. Therefore they had to
verify the consistency of the data with respect to the following selection criteria:
1) Artefacts others than the ones which should have been produced by the signal processing described in [i.2]
e.g. due to the additional amplification required in order to provide a listening level of 79 dB SPL.
2) Inconsistencies within one condition due to the selection of the individual speech samples from the database
for subjective evaluation.
3) Inconsistencies within one condition due to statistical variation of the signal processing described in [i.2]
leading to non consistent judgements within this condition.
ETSI
10 ETSI EG 202 396-3 V1.3.1 (2011-02)
4) Inconsistencies due to ITU-T Recommendation P.56 [i.22] level adjustment process chosen for the complete
files including the background noise.
5) Impact of the different listening levels used in the two databases - the French and the Czech database.
As a result of the experts listening test a set of data was selected which is used for the development of the objective
model.
In the selection process five expert listeners (not native French/Czech speakers) were involved. Their task was not to
produce new judgements, but to check all the samples in the database with respect to the possible artefacts described
above.
A playback system with calibrated headphones was used for the test. The headphones used were Sennheiser HD 600
connected to the HEAD acoustics playback system HPS V. The equalization provided by the headphone manufacturer
was used since this was the one used in the French and Czech test setup.
All samples could be heard by the experts as often as required in order to get final agreement about the applicability of
the data within the terms of reference of the model. There was no limitation in comparing samples to the ones
previously heard.
5.2 Results
In general it could be observed that the 4 seconds sample size chosen in the experiment according to ITU-T
Recommendation P.835 [i.3] lead to a more difficult task even for expert listeners, especially in the case of non
stationary background noises. It is more difficult to identify the nature of the noise itself and then identify in addition
possible impairments introduced by the signal processing or by the network impairments. It is very likely that some
comparatively high standard deviations seen in the data are caused by these effects.
5.3 French database
In general the French database is in line with the ToR except network condition NII. In network condition NII 1 %
packet loss was chosen which is too low for the conditions to be evaluated. Due to the inhomogeneously distributed
packet losses there are conditions where no packet loss is audible up to conditions where 5 out of 6 samples show
packet loss. Furthermore the packet loss may occur during speech as well as during the noise periods. The impact of the
different packet losses is not controlled with respect to their occurrence due to the statistical nature of the packet loss
distribution, even within a set of 6 samples used for evaluating one condition. Since packet loss is clearly audible under
NIII conditions (3 % packet loss) and much better distributed amongst the different samples the NII conditions are not
used within the scope of the objective method. They are either covered by the NI condition (0 % packet loss) or by the
NIII conditions. This results in 144 NII conditions which are not retained for the development of the model.
From the 288 NI and NIII conditions 28 conditions are not retained. The main reasons therefore are:
• Not consistent signal levels due to the amplification process.
• Insufficient S/N, speech almost inaudible.
The individual reasons for the samples of these conditions being not retained can be found in table A.1.
In total 260 out of 432 conditions are used as the reference for the objective model. In other words, 60,2 % of the data
can be used for the model. The distribution of the ratings is between 1,2 and 4,96 MOS for S-/N-/G-MOS.
5.4 Czech database
For every combination of background noise and speaker gender, a single Czech sentence was used (see table 5.1). The
24 Czech listeners had to rate this single sentence, while the French ratings are a mean value of six different sentences
(assessed by 4 listeners each).
ETSI
11 ETSI EG 202 396-3 V1.3.1 (2011-02)
Table 5.1: Sentences from the test corpus chosen for the different conditions
Condition Sentence No.
Lux Car 130kmh Female2 S3
Lux Car 130kmh Male1 S2
Crossroads Female2 S4
Crossroads Male1 S3
Road Noise Female2 S5
Road Noise Male1 S4
Office Noise Female2 S6
Office Noise Male1 S5
Pub Noise Female2 S7
Pub Noise Male1 S6
This leads to a limited representation of the individual background noise conditions especially in the case of time
varying background noises. Furthermore the NII conditions were even more critical in judgement compared to the
French data since either there was no packet loss at all. Or if there was packet loss all listeners rated this particular
packet loss because they all listened to the same sentence for one condition. In the French listening test 6 sentences
were listened for one condition which provided a higher variance of the distributed packet loss.
The listening level variation in the Czech database, preserved from previous database processing adds another degree of
complexity to the problem. The listening levels are generally lower as within the French database and as compared to
the general rules laid down in ITU-Recommendations P.800 [i.4] and P.835 [i.3]. The listening level variation within
the Czech database is up to 16 dB. In the experts tests the following conclusions were drawn:
• The conditions AMR NII and G.722 NII (1 % packet loss) were not selected, because in most cases, the sound
files had too low packet loss. A distinction between and NI and NII conditions is hardly possible.
• The effect of packet loss in the samples should be audible in AMR NIII and G.722 NIII conditions. Because
every single Czech condition consists just of one sentence, the packet loss may not be distributed uniformly in
the sample. Therefore, only samples with at least one packet loss in speech and background noise (before or
after speech) were selected.
• Due to the fact that every Czech sound file has a different level (which depends on codec, noise reduction
algorithm, etc.), a minimum level of 69 dB SPL was set (10 dB below the recommended listening level of
79 dB SPL). All conditions below this limit were not retained.
• Analysis of NI conditions:
a) AMR Codec:
70 conditions were not retained based on the following selection criteria:
1) Too low level (54).
2) Inconsistent BGN level (12).
3) Too low S/N (2).
4) Too low overall level / given listening level not correct (2).
b) G.722 Codec:
19 conditions were not retained based on the following selection criteria:
1) Too low level (15).
2) MOS values irreproducible (4).
c) Selected conditions dependent of BGN: see table 5.2.
ETSI
12 ETSI EG 202 396-3 V1.3.1 (2011-02)
Table 5.2: Selected Czech NI conditions
Selected verification
Total not Total Selected test samples
BGN-Condition samples / no MOS
retained retained / MOS available
available
Lux_Car 17 19 10 9
Crossroads 36 0 0 0
Road 17 1 1 0
Office 14 22 16 6
Pub 5 13 10 3
d) Overall NI acceptance: 48 % of NI conditions are useful (22 % AMR, 65 % G.722).
• Analysis of NIII conditions:
a) AMR Codec:
76 conditions were not retained based on the following selection criteria:
1) Too low level (43).
2) Inconsistent packet loss (33).
b) G.722 Codec:
35 conditions were not retained based on the following selection criteria:
1) Too low level (13).
2) Inconsistent packet loss (22).
c) Selected samples dependent of BGN: see table 5.3.
Table 5.3: Selected Czech NIII conditions
BGN-Condition Total not Total Selected test Selected verification
retained retained samples / MOS samples / no MOS
available available
Lux_Car 30 6 4 2
Crossroads 30 6 5 1
Road 16 2 2 0
Office 24 12 10 2
Pub 11 7 2 5
d) Overall NIII acceptance: 23 % of NIII conditions are useful (16 % AMR, 35 % G.722).
The list of the selected Czech conditions is found in table A.1.
In total 88 conditions out of 432 (20,4 %) are suited to be used in a further step for checking language dependencies.
5.5 General differences between the databases
The most important differences between the French and the Czech database can be summarized as follows:
• The French and Czech listening samples of one condition do not have the same levels. The French sound files
are louder than the Czech ones, in some random tests, the mean of these level differences is given in table A.2,
of EG 202 396-2 [i.2]. This may have lead to different ratings for the Czech samples compared to the French
samples. This has to be regarded especially for further processing of the sound files.
ETSI
13 ETSI EG 202 396-3 V1.3.1 (2011-02)
• For every background noise condition, a single Czech sentence was used (see table 5.1). To quantify the last
point, the correlation between French and Czech ratings (S-, N- and G-MOS) can be calculated. As shown
below, this correlation is very low. It seems that the differences mentioned above are reflected here.
Coefficients of correlation (Pearson's equation) are summarized in table 5.4.
x
MOS Data (Czech)
with:
x
()x − x()y − y
∑ Mean of MOS Data (Czech)
r =
y
2 2
()x − x ()y − y
MOS Data (French)
∑∑
y
Mean of MOS Data (French)
Table 5.4: Comparison of correlation
Only Czech and French selected MOS
Only selected French MOS
Data
Over all available ratings Data (NI and NIII conditions, ratings
(NI and NIII conditions, ratings
(French and Czech, 302 condition each)
reviewed by experts)
reviewed by experts)
(179 selected French conditions)
(59 conditions selected for French and Czech)
S-MOS: 0,703 S-MOS: 0,736 S-MOS: 0,830
N-MOS: 0,816 N-MOS: 0,822 N-MOS: 0,897
G-MOS: 0,668 G-MOS: 0,776 G-MOS: 0,871

As shown in the scatter plots below, a slight correlation for the French-optimized data can be noticed, but for a usable
correlation, the measurement points are distributed too far away from a (virtual) regression line of best fit
(see figures 5.1, 5.3 and 5.5).
If the calculation of the correlation is limited only to the selected data (86 conditions are selected for French and Czech
speech), the correlation increases for all values, especially for the G-MOS data (see figures 5.2, 5.4 and 5.6).

Figure 5.1: Scatter plot of the French data vs. the Czech data for the different conditions,
S-MOS, before experts' selection
ETSI
14 ETSI EG 202 396-3 V1.3.1 (2011-02)

Figure 5.2: Scatter plot of the French data vs. the Czech data, S-MOS, after experts' selection
(only data selected for both languages)

Figure 5.3: Scatter plot of the French data vs. the Czech data for the different conditions,
N-MOS, before experts' selection
ETSI
15 ETSI EG 202 396-3 V1.3.1 (2011-02)

Figure 5.4: Scatter plot of the French data vs. the Czech data, N-MOS, after experts' selection
(only data selected for both languages)

Figure 5.5: Scatter plot of the French data vs. the Czech data for the different conditions, G-MOS,
before experts' selection
ETSI
16 ETSI EG 202 396-3 V1.3.1 (2011-02)

Figure 5.6: Scatter plot of the French data vs. the Czech data, G-MOS, after experts' selection
(only data selected for both languages)
6 Description of the wideband objective test method
6.1 Introduction
The present objective test method is developed in order to calculate objective MOS for speech, noise and the overall
quality of a transmitted signal containing speech and background noise, designated N-MOS, S-MOS and G-MOS in the
following.
The new model is based on an aurally-adequate analysis in order to best cover the listener's perception based on the
previously carried out listening test i.2.
The wideband objective model is applicable for:
• wideband handset and wideband hands-free devices (in sending direction);
• noisy environments (stationary or non-stationary noise);
• different noise reduction algorithms;
• AMR [i.21] and G.722 [i.20] wideband coders;
• VoIP networks introducing packet loss.
NOTE 1: For the NIII conditions jitter was introduced. Finally jitter was observed for less than 2 % of the selected
conditions. The jitter consideration of the new objective method could therefore not be validated on an
appropriate amount of data. Quality impairments typically introduced by different strategies of packet
loss concealment and different adaptive jitter buffer control mechanisms were not considered in the
listening test database and therefore also not in the objective method.
NOTE 2: The method is not applicable for such background situations where speech intelligibility is the major
issue.
Due to the special sample generation process the new method is only applicable for electrically recorded signals. The
quality of terminals can therefore only be determined in sending direction.
ETSI
17 ETSI EG 202 396-3 V1.3.1 (2011-02)
The method was developed by attaching importance to a high reliability. The results of the listening test (selected
conditions, see clause 5) were best modelled. Furthermore mechanisms were implemented to provide high robustness
also for other than the present samples.
Due to the high diversity between the Czech and the French listening test (see clause 5.5) the development of the
objective model is based on the French database being within the ToR and such provides the higher amount of selected
samples. The sample preparation and nomenclatures for the new method are described in clause 6.2.
The calculation of N-MOS, SMOS and GMOS is described in detail in clauses 6.4 to 6.6. Finally clause 6.7 analyses the
results of the new method for the selected French and Czech samples individually and in comparison to each other.
6.2 Speech sample preparation and nomenclature
6.2.1 Speech sample preparation
Based on the data selected in clause 5 an objective model is developed in order to determine:
• the Noise-MOS (N-MOS);
• the Speech-MOS (S-MOS); and
• the "Global"-MOS (G-MOS), the overall quality including speech and background noise.
Different input signals can be accessed during the recording process and subsequently can be used for the calculation of
N-MOS, S-MOS and G-MOS. Beside the signals used in the listening test ("processed signal"), two additional signals
are used as a priori knowledge for the calculation:
1) The "clean speech" signal, which was played back via the artificial mouth at the beginning of the sample
generation process.
2) The "unprocessed signal", which was recorded close to the microphone position of the simulated handset
device / hands-free telephone (see figure 6.1 and [i.2]). Note that no real phone / hands-free device was used.
Phones and handsfree devices were simulated by a free-field microphone and an offline simulation for
filtering, VAD, noise reduction, etc.
Both signals are used in order to determine the degradation of speech and background noise due to the signal processing
as the listeners did during the listening tests.
The sample generation process is shown in figure 6.1.
ETSI
18 ETSI EG 202 396-3 V1.3.1 (2011-02)

NOTE 1: Calibrated for each file with B&K HATS (3.3 ears) to 79 dB SPL ASL (P.56).
NOTE 2: Once calibrated: -26 dBoV resulting to 79 dB SPL measured with a type 3.2 ear (P.57 [i.23]), 5N application force.

Figure 6.1: Sample generation process, indicating "clean speech", "unprocessed speech" and "processed speech"
ETSI
19 ETSI EG 202 396-3 V1.3.1 (2011-02)
The processed signal consists of the unprocessed signal after being processed via noise reduction algorithms, voice
coder, network simulation, etc. This signal was subjectively rated in the previously carried out listening test (see [i.2]
and figure 6.1).
In order to calculate S-MOS, N-MOS and G-MOS, all three signals are required for each sample. The a priori signals
(clean speech and unprocessed) were extracted for each processed signal used in the listening tests.
The following preparation steps are required to be carried out for all three files:
1) The clean and unprocessed speech signals were shortened to 4 seconds in order to match the length of the
processed signal in the listening tests.
2) The signals were time-aligned. This was achieved after pre-processing followed by a cross-correlation
analysis.
NOTE: For samples with an instationary background noise or including packet loss and jitter it should be ensured
that the cross-correlation analysis lead to non-ambiguous results. E.g. by applying further processing
algorithms in order to better separate between speech and noise parts.
The signals are expected to be in a 48 kHz, 16 bit wave format. The clean speech signals are expected to have an Active
Speech Level (ASL, see ITU-T Recommendation P.56 [i.22]) of -4,7 dBPa at the mouth reference point (MRP). For the
unprocessed signal the ASL has to remain unchanged compared to the recording close to the phone's microphone. This
ensures that the influence of phone position and test room is fully obtained. The processed French signals had an ASL
of 79 dB SPL similar to the listening test. The ASL of the Czech processed signals varies between 56 dB SPL and
78 dB SPL and remained unchanged compared to the output of the transmission chain. For further use the speech
signals can have either 79 dB SPL ASL or the originally level after the transmission. Care should be taken that the
corresponding coefficient sets are used (see clauses 6.4 to 6.6).
6.2.2 Nomenclature
In order to provide a consistent nomenclature within the present document, the relevant terms are briefly described in
the following.
The combination of speech sequences, a background noise, a phone type and simulation (filtering, NR level and
aggressiveness), a speech codec and a network scenario leads to one condition in the terms of the present document
and [i.2].
Each condition was generated by processing the clean speech file containing eight sentences per language via the
corresponding scenario, see figure 6.2.

Figure 6.2: Nomenclature (file, condition, sentence)
For the listening tests different parts of the resulting processed files were used. Six of the French sentences per
condition were chosen and assessed by 4 persons each. One of the Czech sentences per condition (randomly, see
table 5.1) was presented to 24 Czech listeners. The resulting auditory S-/N-/G-MOS were averaged in each case
se
...

SLOVENSKI STANDARD
01-junij-2011
.DNRYRVWSUHQRVDJRYRUDLQYHþSUHGVWDYQLKYVHELQ674.DNRYRVWJRYRUDRE
SULVRWQRVWLãXPDR]DGMDGHO3UHQRVãXPDR]DGMD2EMHNWLYQHSUHVNXVQH
PHWRGH
Speech and multimedia Transmission Quality (STQ) - Speech Quality performance in the
presence of background noise - Part 3: Background noise transmission - Objective test
methods
Ta slovenski standard je istoveten z: EG 202 396-3 Version 1.3.0
ICS:
33.040.35 Telefonska omrežja Telephone networks
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.

ETSI Guide
Speech and multimedia Transmission Quality (STQ);
Speech Quality performance
in the presence of background noise
Part 3: Background noise transmission -
Objective test methods
2 ETSI EG 202 396-3 V1.3.1 (2011-02)

Reference
REG/STQ-00167
Keywords
noise, QoS, quality, speech
ETSI
650 Route des Lucioles
F-06921 Sophia Antipolis Cedex - FRANCE

Tel.: +33 4 92 94 42 00  Fax: +33 4 93 65 47 16

Siret N° 348 623 562 00017 - NAF 742 C
Association à but non lucratif enregistrée à la
Sous-Préfecture de Grasse (06) N° 7803/88

Important notice
Individual copies of the present document can be downloaded from:
http://www.etsi.org
The present document may be made available in more than one electronic version or in print. In any case of existing or
perceived difference in contents between such versions, the reference version is the Portable Document Format (PDF).
In case of dispute, the reference shall be the printing on ETSI printers of the PDF version kept on a specific network drive
within ETSI Secretariat.
Users of the present document should be aware that the document may be subject to revision or change of status.
Information on the current status of this and other ETSI documents is available at
http://portal.etsi.org/tb/status/status.asp
If you find errors in the present document, please send your comment to one of the following services:
http://portal.etsi.org/chaircor/ETSI_support.asp
Copyright Notification
No part may be reproduced except as authorized by written permission.
The copyright and the foregoing restriction extend to reproduction in all media.

© European Telecommunications Standards Institute 2011.
All rights reserved.
TM TM TM TM
DECT , PLUGTESTS , UMTS , TIPHON , the TIPHON logo and the ETSI logo are Trade Marks of ETSI registered
for the benefit of its Members.
TM
3GPP is a Trade Mark of ETSI registered for the benefit of its Members and of the 3GPP Organizational Partners.
LTE™ is a Trade Mark of ETSI currently being registered
for the benefit of its Members and of the 3GPP Organizational Partners.
GSM® and the GSM logo are Trade Marks registered and owned by the GSM Association.
ETSI
3 ETSI EG 202 396-3 V1.3.1 (2011-02)
Contents
Intellectual Property Rights . 5
Foreword . 5
1 Scope . 6
2 References . 6
2.1 Normative references . 6
2.2 Informative references . 7
3 Abbreviations . 8
4 Speech signals to be used . 9
5 Selection of the data within the scope of the wideband objective model: Experts evaluation . 9
5.1 Selection process . 9
5.2 Results . 10
5.3 French database . 10
5.4 Czech database . 10
5.5 General differences between the databases . 12
6 Description of the wideband objective test method . 16
6.1 Introduction . 16
6.2 Speech sample preparation and nomenclature . 17
6.2.1 Speech sample preparation . 17
6.2.2 Nomenclature . 19
6.3 Principles of Relative Approach and Δ Relative Approach . 20
6.4 Objective N-MOS. 23
6.4.1 Introduction. 23
6.4.2 Description of N-MOS algorithm . 24
6.4.3 Comparing subjective and objective N-MOS results . 27
6.5 Objective S-MOS . 28
6.5.1 Introduction. 28
6.5.2 Description of S-MOS Algorithm . 28
6.5.3 Comparing Subjective and Objective S-MOS Results . 32
6.6 Objective G-MOS. 32
6.6.1 Description of G-MOS Algorithm . 32
6.6.2 Comparing subjective and objective G-MOS results . 33
6.7 Comparison of the objective method results for Czech and French samples . 34
6.8 Language Dependent Robustness of G-MOS . 38
7 Validation of the Wideband Objective Test Method . 40
7.1 Introduction . 40
7.2 All conditions results analysis . 42
7.2.1 Comparing subjective and objective N-MOS results . 42
7.2.2 Comparing subjective and objective S-MOS results . 43
7.2.3 Comparing Subjective and Objective G-MOS Results . 43
7.3 French Conditions Results Analysed . 44
7.3.1 Comparing Subjective and Objective N-MOS Results . 44
7.3.2 Comparing Subjective and Objective S-MOS Results . 45
7.3.3 Comparing subjective and objective G-MOS results . 45
7.4 Czech conditions results analysis . 46
7.4.1 Comparing subjective and objective N-MOS results . 46
7.4.2 Comparing subjective and objective S-MOS results . 47
7.4.3 Comparing Subjective and Objective G-MOS Results . 47
8 Objective Model for Narrowband Applications . 48
8.1 File pre-processing . 48
8.2 Adaptation of the Calculations . 49
ETSI
4 ETSI EG 202 396-3 V1.3.1 (2011-02)
Annex A: Detailed post evaluation of listening test results . 51
Annex B: Results of PESQ and TOSQA2001 - Analysis of EG 202 396-2 database . 56
Annex C: Comparison of objective MOS versus auditory MOS for the All Data of Training
Period . 63
Annex D: Comparison of objective MOS versus auditory MOS for the Data not used during
the Training Period . 65
Annex E: Regression Coefficients for Czech data . 67
Annex F: Detailed STF 294 subjective and objective validation test results . 68
Annex G: Void . 72
Annex H: Extension of the EG 202 396-3 Speech Quality Test Method to Narrowband:
Adaptation, Training and Validation . 73
Annex I: Validation results of the modified EG 202 396-3 objective speech quality model for
narrowband data . 77
I.1 Introduction . 77
I.2 Description of the Databases . 77
I.3 Collection of the subjective scores . 78
I.4 Differences: HEAD acoustics training database vs. France Telecom validation databases . 80
I.5 Results . 81
I.6 Unmapped Results . 81
I.7 Mapped Results . 84
I.7.1 Use of mapping functions . 84
I.8 Conclusions . 90
History . 92

ETSI
5 ETSI EG 202 396-3 V1.3.1 (2011-02)
Intellectual Property Rights
IPRs essential or potentially essential to the present document may have been declared to ETSI. The information
pertaining to these essential IPRs, if any, is publicly available for ETSI members and non-members, and can be found
in ETSI SR 000 314: "Intellectual Property Rights (IPRs); Essential, or potentially Essential, IPRs notified to ETSI in
respect of ETSI standards", which is available from the ETSI Secretariat. Latest updates are available on the ETSI Web
server (http://webapp.etsi.org/IPR/home.asp).
Pursuant to the ETSI IPR Policy, no investigation, including IPR searches, has been carried out by ETSI. No guarantee
can be given as to the existence of other IPRs not referenced in ETSI SR 000 314 (or the updates on the ETSI Web
server) which are, or may be, or may become, essential to the present document.
Foreword
This ETSI Guide (EG) has been produced by ETSI Technical Committee Speech and multimedia Transmission Quality
(STQ).
The present document is a deliverable of ETSI Specialized Task Force (STF) 294 entitled: "Improving the quality of
eEurope wideband speech applications by developing a performance testing and evaluation methodology for
background noise transmission".
The present document is part 3 of a multi-part deliverable covering Speech and multimedia Transmission Quality
(STQ); speech quality performance in the presence of background noise, as identified below:
Part 1: "Background noise simulation technique and background noise database";
Part 2: "Background noise transmission - Network simulation - Subjective test database and results";
Part 3: "Background noise transmission - Objective test methods".
ETSI
6 ETSI EG 202 396-3 V1.3.1 (2011-02)
1 Scope
The present document aims to identify and define testing methodologies which can be used to objectively evaluate the
performance of narrowband and wideband terminals and systems for speech communication in the presence of
background noise.
Background noise is a problem in mostly all situations and conditions and need to be taken into account in both,
terminals and networks. The present document provides information about the testing methods applicable to objectively
evaluate the speech quality in the presence of background noise. The present document includes:
• The description of the experts post evaluation process chosen to select the subjective test data being within the
scope of the objective methods.
• The results of the performance evaluation of the currently existing methods described in ITU-T
Recommendation P.862 [i.16], [i.17] and in TOSQA2001 [i.19] which is chosen for the evaluation of terminals
in the framework of ETSI VoIP speech quality test events [i.8], [i.9], [i.10] and [i.11].
• The method which is applicable to objectively determine the different parameters influencing the speech
quality in the presence of background noise taking into account:
- the speech quality;
- the background noise transmission quality;
- the overall quality.
• The document is to be used in conjunction with:
- EG 202 396-1 [i.1] which describes a recording and reproduction setup for realistic simulation of
background noise scenarios in lab-type environments for the performance evaluation of terminals and
communication systems.
- EG 202 396-2 [i.2] which describes the simulation of network impairments and how to simulate realistic
transmission network scenarios and which contains the methodology and results of the subjective scoring
for the data forming the basis of the present document.
- French speech sentences as defined in ITU-T Recommendation P.501 [i.13] for wideband and English
speech sentences as defined in ITU-T Recommendation P.501 [i.13] for narrowband.
2 References
References are either specific (identified by date of publication and/or edition number or version number) or
non-specific. For specific references, only the cited version applies. For non-specific references, the latest version of the
reference document (including any amendments) applies.
Referenced documents which are not found to be publicly available in the expected location might be found at
http://docbox.etsi.org/Reference.
NOTE: While any hyperlinks included in this clause were valid at the time of publication ETSI cannot guarantee
their long term validity.
2.1 Normative references
The following referenced documents are necessary for the application of the present document.
Not applicable.
ETSI
7 ETSI EG 202 396-3 V1.3.1 (2011-02)
2.2 Informative references
The following referenced documents are not necessary for the application of the present document but they assist the
user with regard to a particular subject area.
[i.1] ETSI EG 202 396-1: "Speech and multimedia Transmission Quality (STQ); Speech quality
performance in the presence of background noise; Part 1: Background noise simulation technique
and background noise database".
[i.2] ETSI EG 202 396-2: "Speech Processing, Transmission and Quality Aspects (STQ); Speech
Quality performance in the presence of background noise; Part 2: Background Noise Transmission
- Network Simulation - Subjective Test Database and Results".
[i.3] ITU-T Recommendation P.835: "Subjective test methodology for evaluating speech
communication systems that include noise suppression algorithm".
[i.4] ITU-T Recommendation P.800: "Methods for subjective determination of transmission quality".
[i.5] ITU-T Recommendation P.831: "Subjective performance evaluation of network echo cancellers".
[i.6] Genuit, K.: "Objective Evaluation of Acoustic Quality Based on a Relative Approach", InterNoise
'96, Liverpool, UK.
[i.7] ITU-T Recommendation SG 12 Contribution 34: "Evaluation of the quality of background noise
transmission using the "Relative Approach"".
[i.8] ETSI 2nd Speech Quality Test Event: "Anonymized Test Report", ETSI Plugtests, HEAD
acoustics, T-Systems Nova.
NOTE: Available at: http://www.etsi.org/WebSite/OurServices/Plugtests/History.aspx.
Also available as ETSI TR 102 648-3.
[i.9] ETSI 3rd Speech Quality Test Event: "Anonymized Test Report "IP Gateways"".
NOTE: Available at: http://www.etsi.org/WebSite/OurServices/Plugtests/History.aspx.
[i.10] ETSI 3rd Speech Quality Test Event: "Anonymized Test Report "IP Phones"".
[i.11] ETSI 4th Speech Quality Test Event: "Anonymized Test Report "IP Gateways and IP Phones"".
NOTE: Available at: http://www.etsi.org/WebSite/OurServices/Plugtests/History.aspx.
[i.12] F. Kettler, H.W. Gierlich, F. Rosenberger: "Application of the Relative Approach to Optimize
Packet Loss Concealment Implementations", DAGA, March 2003, Aachen, Germany.
[i.13] ITU-T Recommendation P.501: "Test Signals for Use in Telephonometry".
[i.14] R. Sottek, K. Genuit: "Models of Signal Processing in human hearing", International Journal of
Electronics and Communications (AEÜ) vol. 59, 2005, p. 157-165.
NOTE: Available at: http://www.elsevier.de/aeue.
[i.15] SAE International - Document 2005-01-2513: "Tools and Methods for Product Sound Design of
Vehicles" R. Sottek, W. Krebber, G. Stanley.
[i.16] ITU-T Recommendation P.862: "Perceptual evaluation of speech quality (PESQ): An objective
method for end-to-end speech quality assessment of narrowband telephone networks and speech
codecs".
[i.17] ITU-T Recommendation P.862.1: "Mapping function for transforming P.862 raw result scores to
MOS-LQO".
[i.18] ITU-T Recommendation P.862.2: "Wideband extension to Recommendation P.862 for the
assessment of wideband telephone networks and speech codecs".
ETSI
8 ETSI EG 202 396-3 V1.3.1 (2011-02)
[i.19] ITU-T Recommendation SG 12 Contribution 19: "Results of objective speech quality assessment
of wideband speech using the Advanced TOSQA2001".
[i.20] ITU-T Recommendation G.722: "7 kHz audio-coding within 64 kbit/s".
[i.21] ITU-T Recommendation G.722.2: "Wideband coding of speech at around 16 kbit/s using Adaptive
Multi-Rate Wideband (AMR-WB)".
[i.22] ITU-T Recommendation P.56: "Objective measurement of active speech level".
[i.23] ITU-T Recommendation P.57: "Artificial ears".
[i.24] M. Spiegel: "Theory and problems of statistics", McGraw Hill, 1998.
[i.25] R.A. Fisher: "Statistical methods and scientific inference", Oliver and Boyd, 1956.
[i.26] M. Kendall: "Rank correlation methods", Charles Griffin & Company Limited, 1948.
[i.27] Sottek, R.: "Modelle zur Signalverarbeitung im menschlichen Gehör, PHD thesis RWTH Aachen,
1993".
[i.28] ITU-T Recommendation P.830: "Subjective performance assessment of telephone-band and
wideband digital codecs".
[i.29] ITU-T contribution COM 12-117, Study Period 1997-2000: "Report of the question 13/12
rapporteur's meeting (Solothurn, Germany, 6-10 March 2000)".
[i.30] ANSI S1.1-1986 (ASA 65-1986), "Specifications for Octave-Band and Fractional-Octave-Band
Analog and Digital Filters", 1993.
3 Abbreviations
For the purposes of the present document, the following abbreviations apply:
ACR Absolute Comparison Rating
AMR Adaptive MultiRate
ASL Active Speech Level
NOTE: According to ITU-T Recommendation P.56 [i.22].
BGN BackGround Noise
CDF Cumulative Density Function
DB Data Base
dB SPL Sound Pressure Level re 20 µPa in dB
G-MOS Global MOS
NOTE: MOS related to the overall sample.
HP HighPass
IP Internet Protocol
IRS Intermediate Reference System
ITU International Telecommunication Union
ITU-T Telecom Standardization Body of ITU
MOS Mean Opinion Score
MOS-LQSN Mean Opinion Score - Listening Quality Subjective Noise
MRP Mouth Reference Point
NI Network I conditions
NII Network II conditions
NIII Network III conditions
NB NarrowBand
N-MOS Noise MOS
NOTE: MOS related to the noise transmission only.
ETSI
9 ETSI EG 202 396-3 V1.3.1 (2011-02)
NR Noise Reduction
NR (filter) Noise Reduction (filter)
PESQ Perceptual Evaluation of Speech Quality
PLC Packet Loss Concealment
RCV ReCeiVe
RMSE Random Mean Square Error
S-MOS Speech MOS
NOTE: MOS related to the speech signal only.
SNR Signal to Noise Ratio
STF Specialized Task Force
TMOS TOSQA Mean Opinion Score
TOR Terms Of Reference
TOSQA Telecommunication Objective Speech Quality Assessment
VAD Voice Activity Detection
VoIP Voice over IP
WB WideBand
4 Speech signals to be used
As with any objective model, the prediction of speech quality depends on the conditions under which the model was
tested and validated (see clauses 6.1 and 8). This dependency also applies to the speech material used in conjunction
with the objective model.
The wideband version of the model uses French speech sentences. The near end speech signal (clean speech signal)
consists of 8 sentences of speech (2 male and 2 female talkers, 2 sentences each). Appropriate speech samples can be
taken from ITU-T Recommendation P.501 [i.13].
The narrowband version of the model uses English speech sentences. The near end speech signal (clean speech signal)
consists of 8 sentences of speech (2 male and 2 female talkers, 2 sentences each). Appropriate speech samples can be
taken from ITU-T Recommendation P.501 [i.13].
5 Selection of the data within the scope of the
wideband objective model: Experts evaluation
5.1 Selection process
The aim of the selection process was to identify those data in the databases described in EG 202 396-2 [i.2] which are
consistent with the scope of the objective models to be studied within the present document.
The experts were selected on the based on the definition found in e.g. ITU-T Recommendation P.831 [i.5]: experts are
experienced in subjective testing. Experts are able to describe an auditory event in detail and are able to separate
different events based on specific impairments. They are able to describe their subjective impressions in detail. They
have a background in technical implementations of noise reduction systems and transmission impairments and do have
detailed knowledge of the influence of particular implementations on subjective quality.
Their task was to select the relevant conditions within the scope of the model to be developed. Therefore they had to
verify the consistency of the data with respect to the following selection criteria:
1) Artefacts others than the ones which should have been produced by the signal processing described in [i.2]
e.g. due to the additional amplification required in order to provide a listening level of 79 dB SPL.
2) Inconsistencies within one condition due to the selection of the individual speech samples from the database
for subjective evaluation.
3) Inconsistencies within one condition due to statistical variation of the signal processing described in [i.2]
leading to non consistent judgements within this condition.
ETSI
10 ETSI EG 202 396-3 V1.3.1 (2011-02)
4) Inconsistencies due to ITU-T Recommendation P.56 [i.22] level adjustment process chosen for the complete
files including the background noise.
5) Impact of the different listening levels used in the two databases - the French and the Czech database.
As a result of the experts listening test a set of data was selected which is used for the development of the objective
model.
In the selection process five expert listeners (not native French/Czech speakers) were involved. Their task was not to
produce new judgements, but to check all the samples in the database with respect to the possible artefacts described
above.
A playback system with calibrated headphones was used for the test. The headphones used were Sennheiser HD 600
connected to the HEAD acoustics playback system HPS V. The equalization provided by the headphone manufacturer
was used since this was the one used in the French and Czech test setup.
All samples could be heard by the experts as often as required in order to get final agreement about the applicability of
the data within the terms of reference of the model. There was no limitation in comparing samples to the ones
previously heard.
5.2 Results
In general it could be observed that the 4 seconds sample size chosen in the experiment according to ITU-T
Recommendation P.835 [i.3] lead to a more difficult task even for expert listeners, especially in the case of non
stationary background noises. It is more difficult to identify the nature of the noise itself and then identify in addition
possible impairments introduced by the signal processing or by the network impairments. It is very likely that some
comparatively high standard deviations seen in the data are caused by these effects.
5.3 French database
In general the French database is in line with the ToR except network condition NII. In network condition NII 1 %
packet loss was chosen which is too low for the conditions to be evaluated. Due to the inhomogeneously distributed
packet losses there are conditions where no packet loss is audible up to conditions where 5 out of 6 samples show
packet loss. Furthermore the packet loss may occur during speech as well as during the noise periods. The impact of the
different packet losses is not controlled with respect to their occurrence due to the statistical nature of the packet loss
distribution, even within a set of 6 samples used for evaluating one condition. Since packet loss is clearly audible under
NIII conditions (3 % packet loss) and much better distributed amongst the different samples the NII conditions are not
used within the scope of the objective method. They are either covered by the NI condition (0 % packet loss) or by the
NIII conditions. This results in 144 NII conditions which are not retained for the development of the model.
From the 288 NI and NIII conditions 28 conditions are not retained. The main reasons therefore are:
• Not consistent signal levels due to the amplification process.
• Insufficient S/N, speech almost inaudible.
The individual reasons for the samples of these conditions being not retained can be found in table A.1.
In total 260 out of 432 conditions are used as the reference for the objective model. In other words, 60,2 % of the data
can be used for the model. The distribution of the ratings is between 1,2 and 4,96 MOS for S-/N-/G-MOS.
5.4 Czech database
For every combination of background noise and speaker gender, a single Czech sentence was used (see table 5.1). The
24 Czech listeners had to rate this single sentence, while the French ratings are a mean value of six different sentences
(assessed by 4 listeners each).
ETSI
11 ETSI EG 202 396-3 V1.3.1 (2011-02)
Table 5.1: Sentences from the test corpus chosen for the different conditions
Condition Sentence No.
Lux Car 130kmh Female2 S3
Lux Car 130kmh Male1 S2
Crossroads Female2 S4
Crossroads Male1 S3
Road Noise Female2 S5
Road Noise Male1 S4
Office Noise Female2 S6
Office Noise Male1 S5
Pub Noise Female2 S7
Pub Noise Male1 S6
This leads to a limited representation of the individual background noise conditions especially in the case of time
varying background noises. Furthermore the NII conditions were even more critical in judgement compared to the
French data since either there was no packet loss at all. Or if there was packet loss all listeners rated this particular
packet loss because they all listened to the same sentence for one condition. In the French listening test 6 sentences
were listened for one condition which provided a higher variance of the distributed packet loss.
The listening level variation in the Czech database, preserved from previous database processing adds another degree of
complexity to the problem. The listening levels are generally lower as within the French database and as compared to
the general rules laid down in ITU-Recommendations P.800 [i.4] and P.835 [i.3]. The listening level variation within
the Czech database is up to 16 dB. In the experts tests the following conclusions were drawn:
• The conditions AMR NII and G.722 NII (1 % packet loss) were not selected, because in most cases, the sound
files had too low packet loss. A distinction between and NI and NII conditions is hardly possible.
• The effect of packet loss in the samples should be audible in AMR NIII and G.722 NIII conditions. Because
every single Czech condition consists just of one sentence, the packet loss may not be distributed uniformly in
the sample. Therefore, only samples with at least one packet loss in speech and background noise (before or
after speech) were selected.
• Due to the fact that every Czech sound file has a different level (which depends on codec, noise reduction
algorithm, etc.), a minimum level of 69 dB SPL was set (10 dB below the recommended listening level of
79 dB SPL). All conditions below this limit were not retained.
• Analysis of NI conditions:
a) AMR Codec:
70 conditions were not retained based on the following selection criteria:
1) Too low level (54).
2) Inconsistent BGN level (12).
3) Too low S/N (2).
4) Too low overall level / given listening level not correct (2).
b) G.722 Codec:
19 conditions were not retained based on the following selection criteria:
1) Too low level (15).
2) MOS values irreproducible (4).
c) Selected conditions dependent of BGN: see table 5.2.
ETSI
12 ETSI EG 202 396-3 V1.3.1 (2011-02)
Table 5.2: Selected Czech NI conditions
Selected verification
Total not Total Selected test samples
BGN-Condition samples / no MOS
retained retained / MOS available
available
Lux_Car 17 19 10 9
Crossroads 36 0 0 0
Road 17 1 1 0
Office 14 22 16 6
Pub 5 13 10 3
d) Overall NI acceptance: 48 % of NI conditions are useful (22 % AMR, 65 % G.722).
• Analysis of NIII conditions:
a) AMR Codec:
76 conditions were not retained based on the following selection criteria:
1) Too low level (43).
2) Inconsistent packet loss (33).
b) G.722 Codec:
35 conditions were not retained based on the following selection criteria:
1) Too low level (13).
2) Inconsistent packet loss (22).
c) Selected samples dependent of BGN: see table 5.3.
Table 5.3: Selected Czech NIII conditions
BGN-Condition Total not Total Selected test Selected verification
retained retained samples / MOS samples / no MOS
available available
Lux_Car 30 6 4 2
Crossroads 30 6 5 1
Road 16 2 2 0
Office 24 12 10 2
Pub 11 7 2 5
d) Overall NIII acceptance: 23 % of NIII conditions are useful (16 % AMR, 35 % G.722).
The list of the selected Czech conditions is found in table A.1.
In total 88 conditions out of 432 (20,4 %) are suited to be used in a further step for checking language dependencies.
5.5 General differences between the databases
The most important differences between the French and the Czech database can be summarized as follows:
• The French and Czech listening samples of one condition do not have the same levels. The French sound files
are louder than the Czech ones, in some random tests, the mean of these level differences is given in table A.2,
of EG 202 396-2 [i.2]. This may have lead to different ratings for the Czech samples compared to the French
samples. This has to be regarded especially for further processing of the sound files.
ETSI
13 ETSI EG 202 396-3 V1.3.1 (2011-02)
• For every background noise condition, a single Czech sentence was used (see table 5.1). To quantify the last
point, the correlation between French and Czech ratings (S-, N- and G-MOS) can be calculated. As shown
below, this correlation is very low. It seems that the differences mentioned above are reflected here.
Coefficients of correlation (Pearson's equation) are summarized in table 5.4.
x
MOS Data (Czech)
with:
x
()x − x()y − y
∑ Mean of MOS Data (Czech)
r =
y
2 2
()x − x ()y − y
MOS Data (French)
∑∑
y
Mean of MOS Data (French)
Table 5.4: Comparison of correlation
Only Czech and French selected MOS
Only selected French MOS
Data
Over all available ratings Data (NI and NIII conditions, ratings
(NI and NIII conditions, ratings
(French and Czech, 302 condition each)
reviewed by experts)
reviewed by experts)
(179 selected French conditions)
(59 conditions selected for French and Czech)
S-MOS: 0,703 S-MOS: 0,736 S-MOS: 0,830
N-MOS: 0,816 N-MOS: 0,822 N-MOS: 0,897
G-MOS: 0,668 G-MOS: 0,776 G-MOS: 0,871

As shown in the scatter plots below, a slight correlation for the French-optimized data can be noticed, but for a usable
correlation, the measurement points are distributed too far away from a (virtual) regression line of best fit
(see figures 5.1, 5.3 and 5.5).
If the calculation of the correlation is limited only to the selected data (86 conditions are selected for French and Czech
speech), the correlation increases for all values, especially for the G-MOS data (see figures 5.2, 5.4 and 5.6).

Figure 5.1: Scatter plot of the French data vs. the Czech data for the different conditions,
S-MOS, before experts' selection
ETSI
14 ETSI EG 202 396-3 V1.3.1 (2011-02)

Figure 5.2: Scatter plot of the French data vs. the Czech data, S-MOS, after experts' selection
(only data selected for both languages)

Figure 5.3: Scatter plot of the French data vs. the Czech data for the different conditions,
N-MOS, before experts' selection
ETSI
15 ETSI EG 202 396-3 V1.3.1 (2011-02)

Figure 5.4: Scatter plot of the French data vs. the Czech data, N-MOS, after experts' selection
(only data selected for both languages)

Figure 5.5: Scatter plot of the French data vs. the Czech data for the different conditions, G-MOS,
before experts' selection
ETSI
16 ETSI EG 202 396-3 V1.3.1 (2011-02)

Figure 5.6: Scatter plot of the French data vs. the Czech data, G-MOS, after experts' selection
(only data selected for both languages)
6 Description of the wideband objective test method
6.1 Introduction
The present objective test method is developed in order to calculate objective MOS for speech, noise and the overall
quality of a transmitted signal containing speech and background noise, designated N-MOS, S-MOS and G-MOS in the
following.
The new model is based on an aurally-adequate analysis in order to best cover the listener's perception based on the
previously carried out listening test i.2.
The wideband objective model is applicable for:
• wideband handset and wideband hands-free devices (in sending direction);
• noisy environments (stationary or non-stationary noise);
• different noise reduction algorithms;
• AMR [i.21] and G.722 [i.20] wideband coders;
• VoIP networks introducing packet loss.
NOTE 1: For the NIII conditions jitter was introduced. Finally jitter was observed for less than 2 % of the selected
conditions. The jitter consideration of the new objective method could therefore not be validated on an
appropriate amount of data. Quality impairments typically introduced by different strategies of packet
loss concealment and different adaptive jitter buffer control mechanisms were not considered in the
listening test database and therefore also not in the objective method.
NOTE 2: The method is not applicable for such background situations where speech intelligibility is the major
issue.
Due to the special sample generation process the new method is only applicable for electrically recorded signals. The
quality of terminals can therefore only be determined in sending direction.
ETSI
17 ETSI EG 202 396-3 V1.3.1 (2011-02)
The method was developed by attaching importance to a high reliability. The results of the listening test (selected
conditions, see clause 5) were best modelled. Furthermore mechanisms were implemented to provide high robustness
also for other than the present samples.
Due to the high diversity between the Czech and the French listening test (see clause 5.5) the development of the
objective model is based on the French database being within the ToR and such provides the higher amount of selected
samples. The sample preparation and nomenclatures for the new method are described in clause 6.2.
The calculation of N-MOS, SMOS and GMOS is described in detail in clauses 6.4 to 6.6. Finally clause 6.7 analyses the
results of the new method for the selected French and Czech samples individually and in comparison to each other.
6.2 Speech sample preparation and nomenclature
6.2.1 Speech sample preparation
Based on the data selected in clause 5 an objective model is developed in order to determine:
• the Noise-MOS (N-MOS);
• the Speech-MOS (S-MOS); and
• the "Global"-MOS (G-MOS), the overall quality including speech and background noise.
Different input signals can be accessed during the recording process and subsequently can be used for the calculation of
N-MOS, S-MOS and G-MOS. Beside the signals used in the listening test ("processed signal"), two additional signals
are used as a priori knowledge for the calculation:
1) The "clean speech" signal, which was played back via the artificial mouth at the beginning of the sample
generation process.
2) The "unprocessed signal", which was recorded close to the microphone position of the simulated handset
device / hands-free telephone (see figure 6.1 and [i.2]). Note that no real phone / hands-free device was used.
Phones and handsfree devices were simulated by a free-field microphone and an offline simulation for
filtering, VAD, noise reduction, etc.
Both signals are used in order to determine the degradation of speech and background noise due to the signal processing
as the listeners did during the listening tests.
The sample generation process is shown in figure 6.1.
ETSI
18 ETSI EG 202 396-3 V1.3.1 (2011-02)

NOTE 1: Calibrated for each file with B&K HATS (3.3 ears) to 79 dB SPL ASL (P.56).
NOTE 2: Once calibrated: -26 dBoV resulting to 79 dB SPL measured with a type 3.2 ear (P.57 [i.23]), 5N application force.

Figure 6.1: Sample generation process, indicating "clean speech", "unprocessed speech" and "processed speech"
ETSI
19 ETSI EG 202 396-3 V1.3.1 (2011-02)
The processed signal consists of the unprocessed signal after being processed via noise reduction algorithms, voice
coder, network simulation, etc. This signal was subjectively rated in the previously carried out listening test (see [i.2]
and figure 6.1).
In order to calculate S-MOS, N-MOS and G-MOS, all three signals are required for each sample. The a priori signals
(clean speech and unprocessed) were extracted for each processed signal used in the listening tests.
The following preparation steps are required to be carried out for all three files:
1) The clean and unprocessed speech signals were shortened to 4 seconds in order to match the length of the
processed signal in the listening tests.
2) The signals were time-aligned. This was achieved after pre-processing followed by a cross-correlation
analysis.
NOTE: For samples with an instationary background noise or including packet loss and jitter it should be ensured
that the cross-correlation analysis lead to non-ambiguous results. E.g. by applying further processing
algorithms in order to better separate between speech and noise parts.
The signals are expected to be in a 48 kHz, 16 bit wave format. The clean speech signals are expected to have an Active
Speech Level (ASL, see ITU-T Recommendation P.56 [i.22]) of -4,7 dBPa at the mouth reference
...