Audio Bandwidth Extension Using Single Energy Value
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing bandwidth extension techniques for audio signals often result in speech that sounds muffled and has reduced intelligibility due to inaccuracies in estimating the high-band spectral envelope, leading to artifacts and discontinuities in the wideband spectral envelope.
Innovation Solution
A method that uses a single energy value to determine both the spectral envelope shape and energy for out-of-band content, allowing for improved audio quality by estimating the out-of-band energy and using it to synthesize a bandwidth-extended version of the digital audio signal.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If separate estimation of high-band spectral envelope shape and energy is performed, then estimation flexibility is improved, but process complexity increases and discontinuities may occur at band boundaries
Solution Approach 1:
The patent combines the separate estimation processes of spectral envelope shape and energy into a unified estimation framework. By using a single energy value to determine both parameters simultaneously, the method reduces processing complexity while maintaining estimation accuracy and avoiding discontinuities at band boundaries.
Solution Approach 2:
The single energy value serves multiple functions: it determines both the spectral envelope shape and the energy level of the out-of-band content. This multi-functional approach simplifies the processing pipeline by eliminating the need for separate estimation steps while ensuring consistency between shape and energy parameters.
2Measurement precision
If codebook mapping is used to estimate wideband spectral envelope, then high-band spectral envelope shape can be obtained, but estimation errors propagate to high-band energy estimation
Solution Approach 1:
The patent segments the spectral envelope estimation into shape and energy components that are determined independently but consistently through a unified energy value. This segmentation prevents error propagation by ensuring that shape and energy are derived from the same underlying energy measurement rather than through cascaded estimation steps.
3Quantity of substance
If narrow-band speech techniques are used to limit information quantity, then data overhead is reduced, but speech quality becomes muffled and intelligibility decreases
Solution Approach 1:
The patent performs preliminary estimation of the out-of-band energy based on in-band speech characteristics before synthesizing the extended bandwidth content. This preliminary action enables accurate reconstruction of high-frequency components without requiring actual transmission of high-band data, thus maintaining speech quality while using narrow-band transmission.
Data Source
AI summary
One provides (101) a digital audio signal having a corresponding signal bandwidth, and then provides (102) an energy value that corresponds to at least an estimate of out-of-signal bandwidth energy as corresponds to that digital audio signal. One then uses (103) the energy value to simultaneously determine both a spectral envelope shape and a corresponding suitable energy for the spectral envelope shape for out-of-signal bandwidth content as corresponds to the digital audio signal. By one approach, if desired, one then combines (104) (on, for example, a frame by frame basis) the digital audio signal with the out-of-signal bandwidth content to provide a bandwidth extended version of the digital audio signal to be audibly rendered to thereby improve corresponding audio quality of the digital audio signal as so rendered.


