International organisation for standardisation organisation internationale de normalisation
Received National Body Comments and Liaison matters
Yüklə 2,78 Mb.
|
- Bu səhifədəki naviqasiya:
- 7.8Plenary Discussion
7.7Received National Body Comments and Liaison mattersThe NB Comments and Liaison documents for the meeting that require a response are as shown below.
7.8Plenary Discussion7.8.1Audio Reference Code and MPEG SVN Server7.8.2“Round-Trip Coding” and audio conformance and reference software7.9Record of AhG meetings7.10AhG Meeting on SAOC and USAC -- Sunday 1000-18007.10.1Spatial Audio Object CodingOliver Hellmuth, FhG, presented
This overview indicated the topics that will be covered in separate contributions and ones that require more development, and so are not covered in any contribution to this meeting. One topic is presented in this contribution, which is a method for dynamic update of objects (in a teleconferencing application, this would be the number of talkers or locations in a conference call). The conclusion is that the current technology is sufficient to accommodate this, so no action is needed. Leonid Terentiev, FhG, presented
The contribution lists a number of editorial corrections to the FCD text. It is the recommendation of the AhG to incorporate these changes into the FCD text. Jeongil Seo, ETRI, presented
The contribution reviewed the Music 2.0 product, which launched in 2007 in Korea. This has between 4 and 8 tracks which are in stereo so that aspects of spaciousness. ETRI expresses concern that the current limitations in EKS make it insufficient to support the Music 2.0 application. Jeongil Seo, ETRI, presented
The contribution presents a possible technical solution to address the concerns raised in the previous contribution. It suggests that “classic” SAOC mix could be uses as a BGO in the EKS mix. It also gives preliminary evaluation of complexity of the proposed scheme. Leonid Terentiev, FhG, presented
The contribution
For combining EKS and classic SAOC, it proposes to permit an SOAC bitstream to have
The unified EKS/SAOC encoding would first apply EKS processing on the “enhanced” objects to produce a Foreground (stereo) signal and (typically) a residual signal plus a Background (stereo) signal. The latter is processed as a downmix for “regular” SAOC processing, in which any number of “regular” SOAC objects can be manipulated. The Chair suggested that a figure giving a very specific and tutorial example of EKS/SAOC processing should be in the specification. It appears that the concerns of ETRI and LG are addressed by this contribution, but there will be further discussion and the issue will be re-visited mid-week. Finally, the contribution proposes a means to signal a flexible transition between prediction and energy mode for use in upmixing, which can be advantageously used with residual coding. The AhG recommends incorporating this into the FCD text. Jeongil Seo, ETRI, presented
The contribution proposes a container mechanism for carrying the MPEG Surround Multi-channel Background Object in the SAOC bitstream. The container is identical what is currently used for the SAOC residual signal. The presenter noted that the proposal in this contribution is identical to what is proposed in m16653 (the next presentation). Leonid Terentiev, FhG, asked why an explicit bsMPOFlag bit is required since the presence of the container seems to be sufficient to signal that MBO is present. Heiko Purnhagen, Dolby, noted that is it not necessary to indicate the position of the BMP in the bitstream, as it is assumed to be that last stereo element. Leonid Terentiev, FhG, presented
The contribution proposes a container mechanism for carrying the MPEG Surround Multi-channel Background Object in the SAOC bitstream that is identical to what is proposed in the previous contribution. In addition, it proposed to clarify a number of parameters associated with MPEG Surround parameters. The AhG recommends incorporating this into the FCD text. Leonid Terentiev, FhG, presented
The contribution proposes clarifications to the FCD text associated with transcoding of SAOC parameters for MPEG Surround decoding. The Chair suggested that Table 1 which indicates how to map parameters from SAOC to MPS be re-arranged to suggest a signal flow something like this
The AhG recommends incorporating this into the FCD text. Heiko Purnhagen, Dolby, presented
The contribution is a progress report on implementing a low-delay SOAC/MPEG Surround system. It shows subjective test results that indicate that the low-delay implementation has quality comparable to that of the regular system. The low-delay filterbank used in MPEG Surround was exactly the one that is used in low-delay SAOC. The presenter anticipates that the filterbank structure is stable, but other aspects of the decoder may still need to be adapted to get the desired level of performance. It is anticipated that the low-delay flavour of MPEG Surround will be specified as a clause in the SAOC specification. As to next steps, it is suggested that a workplan be drafted that coordinates subjective tests of system performance. Yüklə 2,78 Mb. Dostları ilə paylaş:
|