Directional Loudness Map Based Audio Processing

Simple SummaryContent extracted from patent full text and abstract with AI.

This patent introduces a method and apparatus for analyzing and processing audio, specifically by creating 'directional loudness maps' that show how loudness is distributed across different spatial directions in an audio scene. By extracting spectral and directional information from input signals, the system is able to determine, encode, or decode how sound should be distributed in space, which helps preserve spatial audio quality—especially in compressed or encoded audio formats such as stereo or multichannel audio. This technique also enables more effective objective measurement and optimization of audio quality based on how listeners actually perceive spatial characteristics, such as the position or movement of sounds within the stereo or surround sound field.

Use CasesContent extracted from patent full text and abstract with AI.

• Objective, automated assessment of spatial audio quality in codecs or streaming services to replace or supplement listening tests.
• Spatial audio encoding/decoding for music, movies, and games to enhance or preserve the listener's perception of directionality and spatial cues.
• Adaptive bit allocation and compression for multichannel or object-based audio (e.g., Dolby Atmos, 3D audio) based on the importance of signals in the directional loudness map.
• Format conversion and rescaling audio content (downmixing, upmixing, changing number of channels) while maintaining perceptual quality.
• Rendering and playback optimization for stereo, surround, binaural, or soundbar systems—especially when computational resources are limited (e.g., on mobile devices).
• Efficient transmission and reconstruction of spatial audio scenes, including as side information in audio codecs.

BenefitsContent extracted from patent full text and abstract with AI.

• Improves the perceptual accuracy of spatial audio quality measurement—better predicting listener experience than previous monaural or cue-based metrics.
• Enables more robust encoding and decoding of spatial audio, reducing the loss of spatial cues and image collapses even at low bitrates.
• Optimizes bit distribution and encoding parameters by focusing resources on perceptually important directions or channels, increasing coding efficiency.
• Allows for flexible adaptation of decoding/rendering complexity and format conversion, skipping or prioritizing signals based on their perceptual spatial importance.
• Reduces risk of overfitting and improves generalization of automated quality measurements by using fewer, more perceptually relevant features.
• Facilitates high-quality spatial audio experiences across a range of playback environments and device capabilities.

Technical Classifications (CPCs)

Inventors & Applicants

Patent Abstract

An audio analyzer configured to obtain spectral domain representations of two or more input audio signals. Additionally the audio analyzer is configured to obtain directional information associated with spectral bands of the spectral domain representations and to obtain loudness information associated with different directions as an analysis result. Contributions to the loudness information are determined in dependence on the directional information.