Multisignal audio coding using signal whitening as processing

Patent metadata

Field	Value
Publication number	US-12367883-B2
Application number	US-202017124628-A
Country	US
Kind code	B2
Filing date	Dec 17, 2020
Priority date	Jul 4, 2018
Publication date	Jul 22, 2025
Grant date	Jul 22, 2025

Abstract

A multisignal encoder for encoding at least three audio signals, including: a signal preprocessor for individually preprocessing each audio signal to obtain at least three preprocessed audio signals, wherein the preprocessing is performed so that a preprocessed audio signal is whitened with respect to the signal before preprocessing; an adaptive joint signal processor for performing a processing of the at least three preprocessed audio signals to obtain at least three jointly processed signals or at least two jointly processed signals and an unprocessed signal; a signal encoder for encoding each signal to obtain one or more encoded signals; and an output interface for transmitting or storing an encoded multisignal audio signal including the one or more encoded signals, side information relating to the preprocessing and side information relating to the processing.

First claim

The invention claimed is:

1. Multisignal encoder for encoding at least three audio signals, comprising:
a signal preprocessor for individually preprocessing each audio signal to acquire at least three preprocessed audio signals, wherein the preprocessing is performed so that a preprocessed audio signal is whitened with respect to the signal before preprocessing;
an adaptive joint signal processor for performing a processing of the at least three preprocessed audio signals subsequent to an energy level normalization performed by the adaptive joint signal processor to acquire, in a first alternative, at least three jointly processed signals or, in a second alternative, at least two jointly processed signals and an unprocessed signal,
wherein, for performing the energy level normalization, the adaptive joint signal processor is configured to:
calculate an information on a mean energy of the preprocessed audio signals;
calculate an information on an energy of each preprocessed audio signal,
calculate a broadband energy normalization value for a specific preprocessed audio signal based on the information on the mean energy and the information on the energy of the specific preprocessed audio signal, and
perform a broadband energy normalization of the at least three preprocessed audio signals using the broadband energy normalization values for the specific preprocessed audio signals, wherein each preprocessed audio signal has a normalized energy due to the performing the broadband energy normalization;
wherein the adaptive joint signal processor is configured:
to calculate cross-correlation values for each possible signal pair of the preprocessed signals having the normalized energies of the at least three preprocessed signals;
to select the signal pair having the highest cross-correlation value;
to determine a joint stereo processing mode for the selected signal pair; and
to joint stereo process the selected signal pair in accordance with the determined joint stereo processing mode to obtain a processed signal pair;
a signal encoder for encoding each signal to acquire one or more encoded signals; and
an output interface for transmitting or storing an encoded multisignal audio signal comprising the one or more encoded signals, side information relating to the preprocessing comprising the broadband energy normalization values for the preprocessed audio signals and side information relating to the processing.

2. Multisignal encoder of claim 1 ,
wherein the adaptive joint signal processor is configured to calculate a scaling ratio as the broadband energy normalization value for the specific preprocessed audio signal from the mean energy and the energy of the specific preprocessed audio signal, and
wherein the adaptive joint signal processor is configured for determining a flag indicating whether the scaling ratio is for an upscaling or a downscaling, and wherein the flag for each preprocessed audio signal is comprised by the side information relating to the processing.

3. The multisignal encoder of claim 2 ,
wherein the adaptive joint signal processor is configured to quantize the scaling ratio into the same quantization range irrespective of whether the scaling is an upscaling or a downscaling.

4. Multisignal encoder of claim 1 , wherein the adaptive joint signal processor is configured to apply a cascaded signal pair processing, or wherein the adaptive joint signal processor is configured to apply a non-cascaded signal pair processing,
wherein, in the cascaded signal pair processing, the signals of a processed signal pair are selectable in a further iteration step comprising a calculation of updated cross-correlation values, selecting the signal pair comprising the highest cross-correlation value, the determination of a joint stereo processing mode for the selected signal pair and the joint stereo processing the selected signal pair in accordance with the determined joint stereo processing mode, or
wherein, in the non-cascaded signal pair processing, the signals of a processed signal pair are not selectable in an additional selecting the signal pair comprising the highest cross-correlation value, the determination of a joint stereo processing mode for the selected signal pair, and the joint stereo processing the selected signal pair in accordance with the determined joint stereo processing mode.

5. Multisignal encoder claim 1 ,
wherein the adaptive joint signal processor is configured to determine the signal to be encoded individually as a signal remaining subsequent to a pairwise processing procedure, and
wherein the adaptive joint signal processor is configured to modify an energy normalization applied to the signal before performing the pairwise processing procedure such as reverting, or at least partly reverting the energy normalization applied to the signal before performing the pairwise processing procedure.

6. The multisignal encoder of claim 1 ,
wherein the adaptive joint signal processor is configured for calculating a scaling factor for each signal based on the information on the energy for the specific signal and the information on the mean energy of the at least three preprocessed audio signals,
wherein the adaptive joint signal processor is configured for quantizing the scaling ratio to acquire a quantized scaling ratio value, the quantized scaling ratio value being used for deriving side information for the scaling ratio for each signal comprised by to the encoded multisignal audio signal, and
wherein the adaptive joint signal processor is configured to derive, from the quantized scaling ratio value, a quantized scaling ratio, wherein the preprocessed audio signal is scaled using the quantized scaling ratio before being used for the pairwise processing of the scaled signal with another correspondingly scaled signal.

7. The multisignal encoder of claim 1 ,
wherein the adaptive joint signal processor is configured for calculating normalized inter-signal cross-correlation values for the possible signal pairs in order to decide and select which signal pair comprises the highest degree of similarities and, therefore, is suitable to be selected as a pair for pairwise processing of the at least three preprocessed audio signals,
wherein the normalized cross-correlation values for each signal pair are stored in a cross-correlation vector, and
wherein the adaptive joint signal processor is configured for determining, whether a signal pair selection of one or more previous frames is to be retained or not by com-paring a cross-correlation vector of the previous frame to the cross-correlation vector of the current frame, and wherein the signal pair selection of the previous frame is retained, when a difference between the cross-correlation vector of the current frame and the cross-correlation vector of the previous frame is lower than a predefined threshold.

8. The multisignal encoder of claim 1 ,
wherein the adaptive joint signal processor is configured to apply a non-cascaded signal-pair processing, in which the signals of the processed signal pair are not selectable in a further signal pair processing, wherein the adaptive joint signal processor is configured for selecting the signal pairs based on a cross-correlation between the signal pairs for the pairwise processing, and wherein the pairwise processing of several selected signal pairs is performed in parallel.

9. The multisignal encoder of claim 8 ,
wherein the adaptive joint signal processor is configured to determine, for a selected signal pair, a stereo encoding mode, and wherein, when the stereo encoding mode is determined to be a dual mono mode, the signals involved in this signal pair are at least partly rescaled and indicated as signals to be encode

Assignees

• Fraunhofer Ges Forschung

Inventors

• Fotopoulou Eleni
• Multrus Markus
• Dick Sascha
• Markovic Goran
• Maben Pallavi
• Korse Srikanth
• Bayer Stefan
• Disch Sascha
• Herre Jürgen

Classifications

• G06F3/162

  Interface to dedicated audio devices, e.g. audio drivers, interface to CODECs · CPC title

• G10L19/03 Primary

  Spectral prediction for preventing pre-echo; Temporary noise shaping [TNS], e.g. in MPEG2 or MPEG4 · CPC title

• G10L19/008 Primary

  Multichannel audio signal coding or decoding using interchannel correlation to reduce redundancy, e.g. joint-stereo, intensity-coding or matrixing · CPC title