公开(公告)号：WO1998003037A1

公开(公告)日：1998-01-22

申请号：PCT/EP1997002875

申请日：1997-06-03

Applicant: FRAUNHOFER-GESELLSCHAFT ZUR FÖRDERUNG DER ANGEWANDTEN FORSCHUNG E.V. ,  AT & T LABORATORIES/RESEARCH ,  LUCENT TECHNOLOGIES, BELL LABORATORIES ,  GBUR, Uwe ,  DIETZ, Martin ,  BRANDENBURG, Karlheinz ,  GERHÄUSER, Heinz ,  HERRE, Jürgen ,  QUAKENBUSH, Schuyler

Inventor： FRAUNHOFER-GESELLSCHAFT ZUR FÖRDERUNG DER ANGEWANDTEN FORSCHUNG E.V. ,  AT & T LABORATORIES/RESEARCH ,  LUCENT TECHNOLOGIES, BELL LABORATORIES

IPC: H04S01/00

CPC classification number: H04B1/665 ,  G10L19/008 ,  G10L19/04 ,  H04B1/667 ,  H04B14/068 ,  H04S1/007 ,  H04S3/00

Abstract: When coding and decoding stereophonic spectral values, both the intensity stereo process and a prediction process are used to achieve a high data compression. When an intensity stereo coding is active in a section composed of scaling factor bands (28), prediction for the right channel (R) is deactivated in this area, so that the prediction results are not used to build the coded stereophonic spectral values. The predictor of the right channel (R) is supplied with stereophonic spectral values for this channel which are in turn decoded by an intensity stereo process, so that prediction for the right channel (R) can continue to adapt itself.

Abstract translation: 在编码和立体声音频的解码，以实现高数据压缩，这两种强度立体声处理和预测都可以使用。 如果比例因子频带（28）的截面的强度立体声编码是有效的，该预测是用于右通道（R）在该区域被去激活，从而使预测的结果不用于形成编码立体声音频。 因此，用于右声道的预测（R）可以继续适应，右声道的预测器（R）被供给，反过来，强度立体声解码立体声音频频谱值用于相同。

公开(公告)号：WO1998003036A1

公开(公告)日：1998-01-22

申请号：PCT/EP1997002874

申请日：1997-06-03

Applicant: FRAUNHOFER-GESELLSCHAFT ZUR FÖRDERUNG DER ANGEWANDTEN FORSCHUNG E.V. ,  AT & T LABORATORIES/RESEARCH ,  LUCENT TECHNOLOGIES ,  GBUR, Uwe ,  DIETZ, Martin ,  TEICHMANN, Bodo ,  BRANDENBURG, Karlheinz ,  GERHÄUSER, Heinz ,  HERRE, Jürgen ,  JOHNSTON, John

Inventor： FRAUNHOFER-GESELLSCHAFT ZUR FÖRDERUNG DER ANGEWANDTEN FORSCHUNG E.V. ,  AT & T LABORATORIES/RESEARCH ,  LUCENT TECHNOLOGIES

IPC: H04S01/00

CPC classification number: H04S1/007

Abstract: In a process for coding stereophonic spectral values, the stereophonic spectral values are first grouped into scaling factor bands (28) to which are associated scaling factors. Sections are then formed which consist of at least one scaling factor band (28). The spectral values are coded in at least one section with a coding table associated to the section among a plurality of coding tables associated each to a coding table number. The number of the used coding table is transmitted together with the coded stereophonic spectral values as supplementary information. At least one additional coding table number is provided which does not relate to a coding table but gives information relevant to the section to which the coding table is associated. A process for decoding stereophonic spectral values which have been partially coded by an intensity stereo process and contain supplementary information uses the relevant information given by the additional coding table numbers to undo the coding of the stereophonic spectral values.

Abstract translation: 编码立体声音频的方法，首先执行立体声音频频谱值在其与比例因子有关的比例因子频带（28）的分组。 随后，形成的部分，每个由至少一个比例因子频带（28）的。 频谱值然后至少一部分内编码有多个编码其各自被分配的码簿，作为辅助信息向编码立体声音频被发送的编码表的码簿表的编码的相关部分。 在这里，至少一个额外的代码被提供，其不指编码，但指示其与同一相关信息相关联的部分。 在强度立体声方法，其具有的辅助信息的立体声音频频谱值解码部分编码的方法，使用表示附加Codiertabellennummeren使本Steroaudiospektralwerte的编码反转的相关信息。