Audio Source Separation for Legacy Content Remixing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Legacy audio content is often mixed and lacks original source signals, making it difficult to remix or upmix for devices with more audio channels, and existing techniques result in poor audio quality due to overlapping sound waves without original information.
Innovation Solution
A method and apparatus that receive input audio content, separate mixed audio sources into individual signals and a residual signal, and generate output audio content by mixing these signals based on spatial information and amplitude adjustments, allowing for remixing, upmixing, or downmixing to enhance spatial positioning and loudness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If legacy audio content is mixed without keeping original source signals, then the audio content can be stored and distributed efficiently, but remixing or upmixing becomes difficult and results in poor audio quality
Solution Approach 1:
The patent applies segmentation by separating the mixed audio signal into multiple independent source signals using blind source separation. Instead of treating the mixed audio as a single unit, the system divides it into distinct source components (e.g., vocals, instruments, ambient noise) that can be independently processed, manipulated, and recombined for high-quality remixing and upmixing operations.
2Adaptability or versatility
If techniques are used to separate mixed audio sources without original information, then remixing is possible, but audio quality deteriorates due to overlapping sound waves
Solution Approach 1:
The patent introduces an intermediary residual signal that captures the overlapping and uncertain portions of the audio mixture. This residual signal acts as a mediator between the separated source signals and the final mixed output, allowing the system to preserve information about overlapping sounds that cannot be cleanly separated, thereby maintaining audio quality while enabling remixing versatility.
3Measurement precision
If audio content is separated into individual sources, then spatial positioning and loudness control can be improved, but the complexity of the audio processing increases
Solution Approach 1:
The patent applies dynamics by making the audio processing system adaptive and configurable. The separation and mixing parameters can be dynamically adjusted based on the specific audio content and desired output format. Users can control the degree of separation, spatial positioning, and loudness balance, allowing the system to optimize between quality and complexity depending on the application requirements.
Data Source
AI summary
In method the following is performed: receiving input audio content representing mixed audio sources; separating the mixed audio sources, thereby obtaining separated audio source signals and a residual signal; and generating output audio content by mixing the separated audio source signals and the residual signal.