Spectral Sharpness Control for Low Bit Rate Audio
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Low bit rate audio/speech transform coding methods, such as BandWidth Extension (BWE), often result in low quality due to imprecise spectral harmonic/noise sharpness, which can be over-harmonic or over-noisy, and fail to accurately predict spectral fine structure across frequency bands.
Innovation Solution
A method to control spectral harmonic/noise sharpness by estimating spectral sharpness parameters at the encoder, quantizing and transmitting them, and adjusting the decoded subbands at the decoder to ensure appropriate harmonic and noise components, using a sharpness measuring parameter based on the energy ratio of average to maximum magnitudes, and normalizing energy levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If low bit rate coding is used for bandwidth extension, then bit rate is reduced, but spectral sharpness precision deteriorates causing over-harmonic or over-noisy artifacts
Solution Approach 1:
The patent applies parameter changes by introducing a spectral sharpness control parameter that modifies the spectral fine structure parameters. This parameter adjusts the sharpness of spectral components by controlling the ratio between harmonic and noise components, allowing the system to compensate for the loss of precision inherent in low bit rate coding while maintaining bandwidth extension efficiency.
2Quantity of substance
If spectral fine structure is predicted with limited bits, then bit rate is reduced, but spectral quality deteriorates
Solution Approach 1:
The patent implements feedback by estimating the spectral sharpness parameter from the decoded signal and comparing it with a target sharpness value. Based on this comparison, the system adjusts the spectral fine structure parameters to correct deviations in spectral quality, thereby maintaining high perceptual quality even with limited bit rate allocation for spectral prediction.
3Length of moving object
If bandwidth extension is applied, then frequency range is extended, but spectral sharpness control is lost
Solution Approach 1:
The patent applies preliminary action by pre-defining target spectral sharpness values for different frequency bands before bandwidth extension is performed. These target values serve as reference points that guide the spectral fine structure synthesis process, ensuring that the extended bandwidth maintains appropriate spectral sharpness characteristics without requiring complex real-time control.
Data Source
AI summary
A transmitted data that includes audio data and a transmitted spectral sharpness parameter representing a spectral harmonic/noise sharpness of a plurality of subbands are received. A measured spectral sharpness parameter is estimated from received audio data. The transmitted spectral sharpness parameter is compared with the measured spectral sharpness parameter. A main sharpness control parameter is formed for each of the decoded subbands. The main sharpness control parameter for each of the decoded subbands is analyzed. Ones of the decoded subbands are sharpened if the corresponding main sharpness control indicates that a corresponding subband is not sharp enough, wherein sharpened subbands are formed. Likewise, ones of the decoded subbands are flattened if the corresponding main sharpness control indicates that a corresponding subband is not flat enough, wherein flattened subbands are formed. An energy level of each sharpened subband and each flattened subband is normalized to keep an energy level of each sharpened and/or flattened subband substantially unchanged.