Broadcast Audio Loudness Leveling With Multi-Band Artifact Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing loudness leveling methods in audio signals struggle to maintain consistent long-term loudness while minimizing short-term artifacts, particularly in broadcast television, where sudden volume adjustments can be annoying and inconsistent with regulatory standards.
Innovation Solution
A system and method that combines short-term and long-term loudness leveling modules to adjust audio signals in real-time, using a processor to calculate gain values based on the difference between desired and actual loudness, while removing artifacts and ensuring compliance with measurement standards like ITU-R BS.1770, using equations for rate limiting to prevent quick gain changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If compressor and limiter methods are used for loudness leveling, then short-term loudness fluctuations can be adjusted quickly, but audible artifacts such as breathing or pumping occur
Solution Approach 1:
The audio signal is divided into multiple frequency bands using a filter bank, allowing independent processing of different frequency ranges. This segmentation enables the system to adjust loudness in specific bands without affecting the entire spectrum, thereby reducing audible artifacts while maintaining fast response to short-term fluctuations.
Solution Approach 2:
The patent applies different processing characteristics to different frequency bands based on their specific loudness characteristics. By making the processing quality local to each frequency band rather than uniform across all frequencies, the system can optimize artifact reduction in each band while maintaining overall loudness control effectiveness.
2Ease of operation
If gain adjustment is applied to the whole audio signal at once, then loudness leveling can be achieved, but audible artifacts such as breathing or pumping occur
Solution Approach 1:
Instead of applying gain adjustment to the entire audio signal as a single unit, the patent segments the signal into multiple frequency bands and applies gain adjustment independently to each band. This segmentation approach maintains operational simplicity while eliminating the audible artifacts that result from monolithic gain adjustment.
Solution Approach 2:
The patent transitions from a single-dimension approach (overall gain adjustment) to a multi-dimensional approach by processing different frequency bands independently. This dimensional expansion allows the system to maintain simplicity in control while achieving artifact-free loudness leveling through frequency-selective processing.
3Object-generated harmful factors
If short-term loudness leveling is optimized, then artifacts are minimized, but long-term loudness consistency is not ensured
Solution Approach 1:
The patent implements a dual-timescale processing architecture where short-term loudness leveling operates continuously to minimize artifacts, while a separate long-term loudness measurement and adjustment mechanism ensures consistent average loudness across the entire audio signal. This continuous operation at multiple timescales simultaneously achieves artifact reduction and long-term loudness stability.
Solution Approach 2:
The system incorporates feedback mechanisms that continuously monitor both short-term and long-term loudness characteristics. The long-term loudness measurement provides feedback to adjust the overall gain, ensuring that artifact-free short-term processing does not compromise long-term loudness consistency. This multi-level feedback system maintains both artifact minimization and loudness stability.
Data Source
AI summary
Systems and methods for leveling loudness variation in an audio signal are described. Embodiments use both a perceptual leveling algorithm and a standards-based loudness measure together to minimize audio process artifacts and ensure that the measured loudness of the processed audio is close to a required measure, according to a particular standard measurement of loudness. These systems and methods can be used either offline or in real-time.