Dynamic EQ and Compression for Spectral Profile Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Amateur audio recordings often suffer from unwanted features such as dips and resonances, microphone-related frequency response anomalies, uncontrolled dynamic range, and varying digital levels, which degrade the quality of the recordings.
Innovation Solution
Applying dynamic equalization and compression (DynEQ) to audio signals by building spectral profiles using quantile curves, adjusting frequency-dependent gains to match a target profile, and incorporating noise estimation to enhance signal quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional EQ and DRC are applied separately to audio signals, then frequency response and dynamic range can be adjusted, but the processing complexity increases and the ability to handle dynamic variations across different frequency bands is limited
Solution Approach 1:
The patent combines equalization and dynamic range compression into a single unified processor that simultaneously handles both frequency-dependent gain and dynamic compression. This integration allows the system to adapt to dynamic variations across different frequency bands without requiring separate processing stages, thereby improving adaptability while managing processing complexity through a cohesive architecture.
Solution Approach 2:
The system implements time-varying, frequency-dependent gains that adapt dynamically to the input signal characteristics. By using short-term and long-term signal statistics to modulate the equalization and compression parameters in real-time, the processor can respond to dynamic variations in the audio signal, improving versatility without requiring overly complex fixed-structure processing.
2Measurement precision
If offline mode is used to analyze entire audio signals for accurate statistics, then processing quality improves, but processing time increases
Solution Approach 1:
The system performs a first-pass analysis of the entire audio signal to compute accurate short-term and long-term statistics before applying the actual DynEQ processing. This preliminary statistical analysis enables the offline mode to achieve high measurement precision by understanding the full dynamic range and spectral characteristics of the signal, while the subsequent processing phase can efficiently apply the pre-computed parameters.
Solution Approach 2:
The patent implements both offline and online processing modes that maintain continuous adaptation to signal characteristics. The offline mode provides accurate baseline statistics, while the online mode continues to update statistics in real-time, ensuring that the system maintains high statistical accuracy without requiring complete re-analysis of the entire signal during playback, thus reducing processing time while preserving precision.
3Reliability
If dynamic range compression is applied to balance loudness, then perceived quality improves, but the natural dynamic variation of the audio signal may be reduced
Solution Approach 1:
The system applies frequency-dependent compression ratios and threshold levels that are tailored to specific frequency bands. By allowing different compression characteristics for different frequencies, the system can compress loud passages to improve quality consistency while preserving the natural dynamic variation in frequency regions where it is musically important, thus resolving the contradiction between quality reliability and dynamic range preservation.
Solution Approach 2:
The dynamic compression parameters are continuously adapted based on short-term and long-term signal statistics, allowing the system to apply compression only when and where it improves quality consistency. The time-varying nature of the compression parameters enables the system to preserve natural dynamics in sections where variation is desirable while applying compression in sections where quality consistency is the priority.
Data Source
Figure 1~10
Figure 2~3A
Figure 3B~4
AI summary
Various embodiments are disclosed for (possibly simultaneously) applying EQ and DRC to audio signals. In an embodiment, a method comprises: dividing an input audio signal into n frames, where n is a positive integer greater than one; dividing each frame of the input audio signal into Nb frequency bands, where Nb is a positive integer greater than one; for each frame n: computing an input level of the input audio signal in each band f, resulting in a input audio level distribution for the input audio signal; computing a gain for each band f based at least in part on a mapping of one or more properties of the input audio level distribution to a reference audio level distribution computed from one or more reference audio signals; and applying each computed gain for each band f to each corresponding band f of the input audio signal.