Hybrid Stereo Audio Coding Across Low and High Frequencies

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional stereo audio coding techniques face challenges in achieving optimal bandwidth efficiency, particularly at intermediate bitrates, due to limitations in combining low bandwidth waveform-coding with parametric stereo parameters extracted in the Quadrature Mirror Filters (QMF) domain.

Innovation Solution

A hybrid coding approach is employed, utilizing a decoder that transforms input signals into a sum-and-difference form for lower frequencies and applies parametric stereo coding for higher frequencies, with high frequency reconstruction to enhance audio quality at specific bitrates, and an encoder that encodes signals using a combination of waveform-coding and parametric stereo encoding.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If parametric stereo parameters are extracted in QMF domain to guide MDCT domain stereo coding, then coding efficiency is improved, but it becomes difficult to get the best out of low bandwidth waveform based stereo coding in MDCT domain

Engineering Contradiction:
Improvecoding efficiencyVSAvoidstereo coding quality
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The patent segments the frequency spectrum into two distinct parts: low frequencies (up to crossover frequency) handled by waveform-based stereo coding in MDCT domain, and high frequencies handled by parametric stereo coding using QMF-derived parameters. This segmentation allows each coding method to operate in its optimal frequency range, resolving the contradiction between coding efficiency and stereo coding quality.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different coding strategies to different frequency regions: waveform-based coding with sum-and-difference transformation for low frequencies where human auditory sensitivity is higher, and parametric coding for high frequencies. This local differentiation of coding quality ensures optimal performance across the entire frequency spectrum while maintaining overall coding efficiency.

Inventive Principle:
Principle #3Local quality

2Quantity of substance

If hybrid coding combines low bandwidth waveform-coding with parametric stereo parameters, then bandwidth efficiency is improved, but coding quality at intermediate bitrates deteriorates

Engineering Contradiction:
Improvebandwidth efficiencyVSAvoidcoding quality
Core Design Contradiction:
Quantity of substanceVSManufacturing precision

Solution Approach 1:

The patent changes the domain of parameter extraction from MDCT to QMF domain, and changes the transformation approach for low frequencies from direct parametric coding to waveform-based sum-and-difference coding. These parameter changes enable the system to achieve both bandwidth efficiency through parametric stereo and coding quality through waveform-based approaches at intermediate bitrates.

Inventive Principle:
Principle #35Parameter changes

3Manufacturing precision

If waveform-coded signals are transformed into sum-and-difference form, then audio quality is improved, but computational complexity increases

Engineering Contradiction:
Improveaudio qualityVSAvoidcomputational complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent applies sum-and-difference transformation only to the waveform-coded low frequency signals up to the crossover frequency, rather than transforming the entire frequency spectrum. This segmentation limits the computational overhead to only the necessary frequency range while maintaining audio quality benefits where they are most needed.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4300488B1Stereo audio encoder and decoder
Publication Date: 2026.03.04 DOLBY INTERNATIONAL AB
  • EP4300488B1 patent drawingFigure 1~2
  • EP4300488B1 patent drawingFigure 3
  • EP4300488B1 patent drawingFigure 4

AI summary

The present disclosure provides methods, devices and computer program products for encoding and decoding a stereo audio signal based on an input signal. According to the disclosure, a hybrid approach of using both parametric stereo coding and a discrete representation of the stereo audio signal is used which may improve the quality of the encoded and decoded audio for certain bitrates.