Audio Enhancement Layer Gain Control for Low-Bitrate Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing Code Excited Linear Prediction (CELP) speech coders face challenges in maintaining high-quality speech and audio reproduction at low bit rates, especially for music and non-speech signals, due to model mismatch, leading to degraded audio quality.
Innovation Solution
The method involves generating an enhancement layer within an audio coding system by scaling the core layer output audio with a set of gain values, selecting an optimal gain value based on low error criteria, and transmitting this information to improve audio quality, which includes using Modified Discrete Cosine Transform (MDCT) and perceptual weighting to minimize distortion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If CELP coding is used for speech and audio signals, then compression efficiency is improved, but audio quality deteriorates for music and non-speech signals due to model mismatch
Solution Approach 1:
The audio signal is divided into two separate coding layers: a core layer using CELP for speech-like signals and an enhancement layer using MDCT for music and non-speech signals. This segmentation allows each layer to use the most appropriate coding method for its specific signal type, resolving the model mismatch problem while maintaining compression efficiency.
Solution Approach 2:
The system dynamically changes coding parameters by selecting different coding modes (CELP-only, MDCT-only, or hybrid) based on the signal characteristics. The enhancement layer introduces new parameters (MDCT coefficients, gain values) that are activated only when needed for music signals, allowing parameter optimization without sacrificing compression efficiency for speech signals.
2Manufacturing precision
If enhancement layer is added to improve audio quality, then data rate increases, but transmission efficiency decreases
Solution Approach 1:
The enhancement layer is applied partially - only when the signal contains music or non-speech components that benefit from MDCT coding. For pure speech signals, only the core CELP layer is used. This partial application of the enhancement layer improves audio quality where needed while minimizing the increase in data rate.
Solution Approach 2:
The system performs preliminary analysis of the audio signal to determine its characteristics (speech-like vs. music-like) before selecting the appropriate coding mode. This preliminary action allows the system to avoid unnecessary enhancement layer processing for speech signals, maintaining transmission efficiency while being ready to apply enhancement when needed.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
During operation a multiple channel audio input signal is received and coded to generate a coded audio signal. A balance factor having balance factor components each associated with an audio signal of the multiple channel audio signal is generated. A gain value to be applied to the coded audio signal to generate an estimate of the multiple channel audio signal based on the balance factor and the multiple channel audio signal is determined, with the gain value configured to minimize a distortion value between the multiple channel audio signal and the estimate of the multiple channel audio signal. The representation of the gain value may be output for transmission and/ or storage.