Automatic loudness control in short-form content for broadcasting.

Area of Science:

• Audio Engineering
• Broadcast Technology
• Signal Processing

Background:

• The International Telecommunication Union (ITU) loudness standard (Rec. BS.1770) identified a need for better loudness descriptors for short-form content.
• Sudden loudness jumps in commercials and live inserts can negatively impact audience experience.
• Existing standards may not adequately address the unique challenges of short-form audio content.

Purpose of the Study:

• To propose and evaluate a novel loudness control scheme for broadcast audio.
• To prevent undesirable loudness jumps in short-form content.
• To enhance viewer satisfaction by ensuring consistent audio levels.

Main Methods:

• Developed a short-form content audio detection method using Principal Component Analysis (PCA) and Support Vector Machines (SVM).
• Implemented dynamic range processing based on short-term loudness integrators and Hilbert transformers for maximum loudness control.
• Utilized quality classification metrics for performance assessment.

Main Results:

• The proposed scheme effectively detects short-form audio content.
• Dynamic range processing successfully mitigates loudness jumps, adhering to maximum loudness level criteria.
• Performance assessment validated the scheme's effectiveness through objective metrics and a practical demonstration.

Conclusions:

• The developed loudness control scheme offers a robust solution for managing audio levels in short-form broadcast content.
• This approach enhances the viewing experience by eliminating abrupt loudness variations.
• The integration of PCA, SVM, and signal processing techniques provides an effective framework for broadcast loudness management.