Intelligent Audio Rendering with Dynamic Sound Object Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies for rendering sound scenes struggle to accurately adapt recorded sound scenes to produce alternative rendered sound scenes, especially when sound sources move within the scene, as they fail to effectively track the position and orientation of sound objects relative to the listener.
Innovation Solution
A system and method that utilize a combination of positioning, orientation, and distance blocks to process audio signals from static and portable microphones, adjusting for movement and orientation to correctly render sound objects within a sound scene, allowing for accurate or intentional mispositioning of sound objects in the rendered sound scene.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current technology is used to render sound scenes, then the rendering process can be completed, but the spatial accuracy and ability to track sound object positions relative to the listener deteriorates
Solution Approach 1:
The patent divides the sound scene into multiple independent sound objects, each with its own position, orientation, and audio properties. This segmentation allows individual tracking of each sound object relative to the listener, improving spatial accuracy without requiring complete re-rendering of the entire scene.
Solution Approach 2:
The system dynamically adjusts the position and orientation of sound objects based on listener movement. By continuously updating the spatial relationships between the listener and sound objects, the system maintains accurate tracking even as the listener moves through the environment.
2Loss of information
If multiple microphones are used to record sound scenes, then the audio quality and spatial information improve, but the complexity of processing and rendering the sound scene increases
Solution Approach 1:
The patent extracts spatial information and sound object properties from multi-microphone recordings, separating the essential spatial data from the raw audio signals. This extraction process captures the necessary spatial information while simplifying subsequent rendering operations by pre-processing the spatial relationships.
Solution Approach 2:
The system performs preliminary processing of multi-microphone recordings to establish sound object positions, orientations, and audio properties before rendering. By pre-establishing these spatial relationships, the system reduces the complexity of real-time rendering operations while preserving spatial information.
3Reliability
If the recorded sound scene is rendered exactly as recorded, then the authenticity is maintained, but the adaptability to different listening conditions and intentional mispositioning is lost
Solution Approach 1:
The patent implements dynamic rendering that allows the sound scene to be adapted based on listening conditions and user preferences. The system can switch between accurate reproduction mode (maintaining authenticity) and flexible rendering mode (allowing intentional mispositioning), providing adaptability while preserving the option for faithful reproduction.
Data Source
Figure 1~3
Figure 4A~6B
Figure 7~9
AI summary
A method comprising: automatically applying a selection criterion or criteria to a sound object; if the sound object satisfies the selection criterion or criteria then performing one of correct or incorrect rendering of the sound object; and if the sound object does not satisfy the selection criterion or criteria then performing the other of correct or incorrect rendering of the sound object, wherein correct rendering of the sound object comprises at least rendering the sound object at a correct position within a rendered sound scene compared to a recorded sound scene and wherein incorrect rendering of the sound object comprises at least rendering of the sound object at an incorrect position in a rendered sound scene compared to a recorded sound scene or not rendering the sound object in the rendered sound scene.