Audio Zooming via Spatial Vector Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio scene technologies lack the ability to accurately identify and indicate specific listening positions, resulting in a compromised user experience due to noise-like sound effects and limited meaningful listening positions, requiring users to find suitable positions through trial and error.
Innovation Solution
A method that analyzes audio scenes by dividing them into cells, determining directional vectors for audio sources, combining vectors within cells, and identifying intersection points as zoomable audio points, which are then provided to client devices for selection, enabling users to focus on relevant sound sources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If audio signals from multiple audio sources are recorded and transmitted to reconstruct an audio scene, then the audio scene can be captured and rendered for user listening, but the audio signals become mixed up resulting in noise-like sound effects
Solution Approach 1:
The patent divides the audio scene into multiple cells and further segments the audio signals within each cell based on directional vectors. This segmentation allows the system to separate mixed audio signals by their spatial origins, enabling users to focus on specific audio sources without the noise-like effect of complete mixing.
Solution Approach 2:
The patent applies local quality by providing different audio processing treatments to different spatial regions (cells) of the audio scene. Each cell's audio signals are processed independently with directional vector analysis, allowing meaningful listening experiences at specific locations while maintaining overall scene reconstruction capability.
2Ease of operation
If users select listening positions from reconstructed audio space, then they can choose preferred listening points, but they must find suitable positions through trial and error without guidance
Solution Approach 1:
The patent performs preliminary analysis of the audio scene to pre-identify optimal listening positions (zoomable audio points) before user selection. By calculating directional vectors and identifying intersection points in advance, the system prepares guidance information that helps users immediately select meaningful listening positions without trial and error.
Solution Approach 2:
The patent implements feedback by providing users with information about zoomable audio points and their characteristics. This feedback mechanism guides users to select optimal listening positions by showing them where meaningful audio sources are located, eliminating the need for random trial and error searching.
3Measurement precision
If the audio scene is analyzed to determine zoomable audio points by dividing into cells and calculating directional vectors, then accurate listening positions can be identified, but the processing complexity increases
Solution Approach 1:
The patent reduces processing complexity by segmenting the audio scene into cells and processing each cell independently. This segmentation allows the complex directional vector calculation to be performed on smaller, manageable subsets of audio signals, maintaining precision while reducing overall computational burden through parallel processing potential.
Data Source
AI summary
A method comprising: obtaining a plurality of audio signals originating from a plurality of audio sources in order to create an audio scene; analyzing the audio scene in order to determine zoomable audio points within the audio scene; and providing information regarding the zoomable audio points to a client device for selecting.