| 摘要 |
<p>The invention lies in multi channel audio processing. In order to preserve the spatial audio image of an input signal, parameters which describe the audio scene must be accurately determined. The invention adds a processor that uses an inter-channel prediction model to form an inter-channel parameter. A processor receives at least a first input audio channel and a second input audio channel and uses an inter-channel prediction model to form at least one inter-channel parameter. Parameters such as inter-channel time difference or inter-channel phase difference or leveldifference inter-channel parameters are determined. Analysis, encoding and decoding of the multi-channel audio is disclosed.</p> |
