|
(Enlarged continuation of Question 6/12 studied in 1997-2000)
Type of question
Task-oriented, leading to new and revised ITU-T Recommendations.
Background
During the 1997-2000 Study Period, a new Recommendation P.581 (Use of
head and torso simulator (HATS) for hands-free terminal testing) has been
produced and Recommendation P.340 (Transmission characteristics and speech
quality parameters of hands-free telephones) has been revised, including a
new section on speech processing (mainly Acoustic Echo Canceller) and
defining new parameters, and for some of them limits.
Recommendation P.502 includes test methodologies that can apply for
hands-free testing and Recommendation P. 501 has enlarged the list of test
signals applicable for testing hands-free terminals and speech processing
devices.
Recommendation P.832 (Subjective performance evaluation of hands-free
terminals) defines three subjective methodologies to evaluate the quality of
hands-free terminals in single and double talk situations.
Product oriented Recommendations P.341 and P.342 for digital hands-free
terminals have been revised.
There is a need to produce new product-oriented Recommendations including
hands-free functions as, mobile, IP, conferencing and audiovisual
terminals.
This methodological question should address the
principles for specifying the terminal performance of hands-free
applications, including
- characterization and specification of advanced speech processing
devices (e.g. noise reduction): pertinent parameters and appropriate
limits ; the test method should be define in collaboration with new
Question 6/12,
- influence of hands-free application on test methodology to apply,
- consequences on the test methodology and parameters, expected
limits of multi-device sound pick-up and restitution,
- consequences of the combination of speech processing devices
(including low bit rate coding) on the quality perceived by the user.
Text of the Question
considering
a) that new types of terminals are and will be defined for different
types of applications, including or not video;
b) that signal processing systems are implemented in terminals (echo
cancellation, noise reduction, advanced signal processing for multi channel
pick-up and restitution…),
the following questions should be studied:
- What are the more appropriate test methods to be applied to
characterize the performance of speech processing devices implemented in
terminals? What limits and/or characteristics may be defined to ensure
that the speech processing devices contribute to improve the overall
quality of the speech transmission, specially in double talk situation?
- What could be the recommended test methodology and consequences of
the use of HATS for the test of new types of hands-free terminals
(associated or not with video)?
- How progresses in psycho-acoustic models could be applicable to
HATS for the benefit of hands-free terminals testing ?
- What are the consequences on the speech quality of the combination
of speech processing devices implemented in hands-free terminals? What
characteristics and limits can apply for these combinations ?
Study items
1) Definition of characteristics and test methods for the noise
reduction devices and combination of AEC and noise reduction systems, in
the context of mobile hands-free terminals. Definition of
characteristics for dereverberation systems. Definition of the test
environment appropriate to these measurements.
2) Influence of combinations of different signal processing
equipments on the perceived quality for hands-free terminals.
3) Applications to multichannel sound pick up and restitution (spatialisation,
stereo).
4) Better global approach of binaural summation for HATS and
calculation of RLR, and influence of the localisation of different
acoustic sources on the binaural behaviour of HATS and on the
measurement methodology.
5) Use of HATS for microphone arrays and multi devices sound
reproduction.
6) Specific test methodologies applicable for hand-held, wearable,
lap-top and conference hands-free terminals, and based on the real use
context. Applications for speech only terminals and for audiovisual
terminals.
7) Determination of recommended values for new parameters (mainly for
temporally weighted echo return loss and noise transmission quality) in
single and double talk situations : definition of basic references
for the new P.34X Series recommendations.
8) Room acoustic conditions for hands-free tests (e.g. automotive
environment).
Objectives
New Recommendation P.SPDA (Speech Processing Devices for Acoustic
enhancement).
Revised Recommendation P.581.
Revised Recommendation P.340.
Relationship with other Study Groups
A close relationship shall be maintained with Study Group 15 dealing with
speech processing .
Relationship with other standardization bodies
ETSI STQ, IEEE and TIA TR-41
|
