Volume Leveling Control Using Two-Stage PGC Noise Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing volume leveling units struggle to differentiate between professionally generated content (PGC) noise and user-generated content (UGC) noise, often boosting unwanted environmental noise in UGC while failing to maintain desired loudness levels for PGC.
Innovation Solution
A two-stage noise classifier system is employed to distinguish between PGC and UGC noise, using a noise detector and discriminator to calculate confidence scores and update the volume leveling control signal only when noise reliability is high, ensuring consistent performance by adjusting the dynamic range controller (DRC) gain based on PGC noise confidence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If the volume leveling unit boosts all noise-like content to maintain consistent loudness, then the loudness consistency is improved, but the unwanted environmental noise in UGC is also boosted
Solution Approach 1:
The patent segments noise classification into two distinct stages: first identifying whether noise is present, then determining whether it is PGC or UGC noise. This segmentation allows the system to apply different handling strategies to different noise types, boosting PGC noise while suppressing UGC noise, thereby resolving the contradiction between maintaining loudness consistency and avoiding environmental noise boosting.
Solution Approach 2:
The patent applies different quality treatments to different parts of the audio signal based on their origin. PGC noise segments receive boosting treatment to maintain loudness, while UGC noise segments receive suppression treatment to remove harmful environmental noise. This local differentiation resolves the contradiction by allowing selective noise handling rather than uniform treatment.
2Object-generated harmful factors
If the volume leveling unit suppresses all noise to eliminate environmental noise, then the harmful noise is reduced, but the desired PGC noise is also suppressed
Solution Approach 1:
The two-stage classification system segments noise into PGC and UGC categories, allowing the system to suppress only UGC noise while preserving PGC noise. This segmentation resolves the contradiction by enabling selective suppression rather than blanket noise reduction, maintaining loudness consistency for desired content while eliminating harmful environmental noise.
Solution Approach 2:
The patent applies different suppression levels to different noise types: strong suppression for UGC environmental noise and no suppression (or even boosting) for PGC noise. This local quality differentiation resolves the contradiction by allowing harmful noise suppression without affecting desired content.
3Device complexity
If a single-stage noise classifier is used to simplify the system, then the device complexity is reduced, but the noise differentiation accuracy between PGC and UGC is insufficient
Solution Approach 1:
The patent segments the classification task into two sequential stages: noise detection followed by PGC/UGC discrimination. This segmentation improves accuracy by breaking down the complex single task into two simpler, more specialized sub-tasks, each optimized for its specific function. The first stage identifies noise presence, and the second stage determines noise origin, achieving high differentiation accuracy without requiring an overly complex single-stage system.
Solution Approach 2:
The first-stage noise detector acts as an intermediary that filters and prepares signals for the second-stage PGC/UGC discriminator. This intermediary structure allows the system to handle different types of input signals appropriately, improving overall classification accuracy while maintaining manageable system complexity through modular design.
4Adaptability or versatility
If the volume leveling control signal is updated continuously to adapt to changing audio content, then the adaptability is improved, but the control signal stability deteriorates
Solution Approach 1:
The patent updates the volume leveling control signal based on periodic analysis of audio segments rather than continuous frame-by-frame updates. By analyzing noise reliability ratios over segmented periods and updating control signals at appropriate intervals, the system achieves adaptability to content changes while maintaining stability through controlled update frequency, preventing excessive fluctuations in the control signal.
Data Source
Figure 1~3
AI summary
Volume leveling of an audio signal using a volume leveling control signal. The method comprises determining a noise reliability ratio w(n) as a ratio of noise-like frames over all frames in a current time segment, determining a PGC noise confidence score X PGN (n) indicating a likelihood that professionally generated content, PGC, noise is present in the time segment, and determining, for the time segment, whether the noise reliability ratio is above a predetermined threshold. When the noise reliability ratio is above the predetermined threshold, the volume leveling control signal is updated based on the PGC noise confidence score, and when the noise reliability ratio is below the predetermined threshold, the volume leveling control signal is left unchanged.Volume leveling is improved by preventing boosting of e.g. phone-recorded environmental noise in UGC, while keeping original behavior for other types of content.