International organisation for standardisation organisation internationale de normalisation

1Opening of the meeting

The MPEG Audio Subgroup meeting was held during the 82^nd meeting of WG11, October 22-26, 2007 in Shenzhen, CN. The list of participants is given in Annex A.

2Administrative matters

2.1Approval of previous meeting report

The 81^st Audio Subgroup meeting report was registered as a contribution, and was approved.

2.2Approval of agenda and allocation of contributions

The agenda and schedule for the meeting was discussed, edited and approved. It shows the documents contributed to this meeting and presented to the Audio Subgroup, either in the task groups or in Audio plenary. The Chair brought relevant documents from Requirements, Systems and MDS to the attention of the group. It was revised in the course of the week to reflect the progress of the meeting, and the final version is shown in Annex B.

2.3Task Groups

Task groups were convened for the duration of the MPEG meeting, as shown in . Results of task group activities are reported below.

2.4Communications from the Chair

The Chair summarised the issues raised at the Sunday evening Chair’s meeting, proposed task groups for the week, and proposed agenda items for discussion in Audio plenary.

2.5Joint meetings

There were no joint meetings with Audio over the course of the week.

2.6Received National Body Comments and Liaison matters

The NB Comments and Liaison documents for the meeting that require a response are as shown below.

No.	Title	Response by
14877	USNB Contribution: Soundfield reproduction technology	S. Quackenbush
14878	USNB Contribution: Profiles for ALS that meet industry needs	S. Quackenbush
14789	Liaison Statement from ITU-T SG 16 [SC 29 N 8677]	None
14791	IEC NP: 100/1269/NP [SC 29 N 8693]	A. Taleb, S. Quackenbush
14910	Liaison Statement from ITU-T SG 16 [SC 29 N 8817]	S. Quackenbush
14911	Liaison Statement from 3GPP2 [SC 29 N 8819]	None
14790	IEC CD 62516 [SC 29 N 8678]	K. Kjörling, E. Oh
14862	Swedish NB comments on IEC CD 62516 [m14790 (SC 29 N 8678)]	K. Kjörling
14865	German NB comments on IEC CD 62516 [m14790 (SC 29 N 8678)]	K. Kjörling

3Record of AhG meetings

3.1AhG Meeting AAC-ELD Sunday 0900-1300

Pierrick Phillip, France Telecom, presented

14978

Pierrick Philippe David Virette

Listening test results on block Switching Core Experiment for ELD-AAC

This gave information on listening tests specified in the workplan and additional test results.

FhG-BS vs FhG (12 listeners)

• 
  Overall – no statistical difference
  
• 
  Per item – 1 better with BS
  

FT-BS vs FT (12 listeners)

• 
  Overall – FT-BS better than FT
  
• 
  Per item – 4 better with BS, none worse
  

FT vs FhG (10 listeners)

• 
  Overall – no difference
  
• 
  Per item – 1 better for FT, 2 worse
  

FT-BS vs FhG-BS (8 listeners)

• 
  Overall – FT-BS better
  
• 
  Per item – 3 better with BS, none worse
  

FT-BS vs FhG-BS (9 listeners, at Philips)

• 
  Overall – FT-BS better than FhG-BS
  
• 
  Per item – 3 better with FT-BS, 1 worse
  

FT-BS vs FhG (11 listeners)

• 
  Overall – FT-BS better than FhG
  
• 
  Per item – 3 better with BS, none worse
  

Kristopher K, Coding Technologies, presented

14984

Fredrik Henn Per Ekstrand

CT cross-check of FT and FhG versions of block-switching for AAC-ELD

This gave information on listening tests specified in the workplan and additional test results.

FT-BS vs FT (9 listeners)

• 
  Overall – FT-BS better than FhG
  
• 
  Per item – 5 better with BS, none worse
  

FhG-BS vs FhG (9 listeners)

• 
  Overall – no difference
  
• 
  Per item – 5 better with BS, none worse
  

FhG vs FT (9 listeners)

• 
  Overall – FhG is better
  
• 
  Per item – 5 better with FhG, none worse
  

FT-BS vs FhG (9 listeners)

• 
  Overall – FT-BS is better
  
• 
  Per item – 2 better with FT-BS, none worse
  

FhG-BS vs FT-BS (6 listeners)

• 
  Overall – FhG-BS is better
  
• 
  Per item – 3 better with FhG-BS, none worse
  

Markus Schnell, FhG, presented

14998

Markus Schnell Ralf Geiger Markus Schmidt Tobias Albert

Cross Check of blockswitching for AAC ELD

This gave information on listening tests specified in the workplan and additional test results.

FT-BS vs FT (14 listeners)

• 
  Overall – FT-BS better
  
• 
  Per item – 8 better with BS, none worse
  

FhG-BS vs FhG (14 listeners)

• 
  Overall – no difference
  
• 
  Per item –none different
  

FhG vs FT (15listeners)

• 
  Per item – 4 better with FhG, none worse
  

FT-BS vs FhG (10 listeners)

• 
  Per item – 2 better with FT-BS, none worse
  

FhG-BS vs FT-BS (10 listeners)

• 
  Per item – 3 better with FT-BS, 2 worse
  

Henney Oh, LG, presented

15010

Henney Oh Yang-Wong Jung Dong-Geum Lee Hong-Goo Kang

Listening Test Reports on Block Switching CE for AAC-ELD

This gave information on listening tests specified in the workplan and additional test results. This reports on tests conducted at LG and Yonsei University.

FT-BS vs FT (11 listeners)

• 
  Overall – FT-BS better
  
• 
  Per item – 3 better with BS, none worse
  

FhG-BS vs FhG (11 listeners)

• 
  Overall – BS is better
  
• 
  Per item –none different
  

FhG vs FT (11 listeners)

• 
  Overall – FhG is better
  
• 
  Per item – 2 better with FhG, none worse
  

FT-BS vs FhG (10 listeners)

• 
  Overall – BS is better
  
• 
  Per item – 2 better with BS, none worse
  

FhG-BS vs FT-BS (10 listeners)

• 
  Overall – no difference
  
• 
  Per item – 1 better with FT-BS, none worse
  

Discussion

Pierrick Phillip, France Telecom, pooled all test results from all test sites to obtain the following graphs.

Based on the test results, Werner Oomen, Philips, constructed the following picture. It shows the systems under test, and the arrows between systems are labelled with the number of test items for which the listening test results showed a statistically significant improvement in performance. The green color is taken from the pooled test results, other colors are results from the individual test sites.

All experts in the AhG agree with the following:

• 
  When pooling all data, FT-BS was better than FT for 8 items and FhG-BS was better than FhG for no items at the 95% level of significance. In terms of the Workplan from the last meeting, N9238 Workplan for AAC-ELD, this level of performance does not trigger any agreed upon action.
  

The Chair asked for a show of hands on the following positions:

Those who had a strong position in support of the BS technology	10
Those who had a strong position that was not in support of the BS technology	5
Those who felt that the BS technology provided quality improvement	14
Those who felt that the BS technology did not show quality improvement	5

Recommendations

The AhG recommends that the Audio Subgroup

• 
  Continue to discuss the points of view captured by the statements above.
  
• 
  That more than audio quality must be considered for adoption of the Core Experiment technology.