Audio Encoder Variable Warping Filter Low Bitrate
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio coding technologies fail to efficiently encode both general audio and speech signals at low bitrates, as general audio coders are not optimized for speech signals and vice versa, due to differences in spectral shaping and frequency resolution requirements.
Innovation Solution
An audio encoding system with a pre-filter having a variable warping characteristic, controlled by a time-varying signal, allows for seamless integration of different coding algorithms within a single framework, using a first algorithm for specific signal patterns and a second algorithm for general audio, with a post-filter performing inverse warping to shape noise according to the masking threshold.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If LPC-based speech coding is used, then speech signals can be efficiently coded at very low bit rates, but general music signals cannot be effectively processed due to inability to flexibly shape spectral envelope
Solution Approach 1:
The patent applies dynamics by making the warping characteristic time-varying rather than fixed. The warping factor is dynamically adjusted based on the analyzed audio signal characteristics, allowing the system to adapt between speech-like and music-like signals in real-time. This enables a single coder to efficiently handle both speech and music by changing its spectral shaping behavior according to the input signal type.
Solution Approach 2:
The patent changes the warping factor parameter dynamically based on signal analysis. By adjusting this key parameter, the system transforms the frequency response characteristics to match different signal types - using higher warping factors for speech signals to enhance low-frequency resolution and lower warping factors for music signals to preserve overall spectral shape, thus resolving the contradiction between specialized efficiency and general adaptability.
2Manufacturing precision
If perceptual audio coding with fixed warping is used, then spectral shaping according to masking threshold is achieved, but computational complexity increases and adaptability to different signal types is reduced
Solution Approach 1:
The system dynamically adjusts the warping factor based on real-time signal analysis rather than using a fixed complex spectral shaping process. This dynamic approach simplifies the overall system architecture by using a single adjustable parameter to achieve the desired spectral shaping, reducing computational complexity while maintaining adaptability to different signal types and preserving masking threshold compliance.
Solution Approach 2:
Instead of using complex fixed spectral shaping algorithms, the patent changes the warping factor parameter to achieve spectral shaping. This parameter-based approach is computationally simpler than fixed complex filtering while still achieving the necessary spectral distortion shaping for perceptual coding, thus reducing device complexity while maintaining manufacturing precision.
3Measurement precision
If high warping characteristic is applied, then frequency resolution is improved for speech signals, but audio quality deteriorates for general music signals
Solution Approach 1:
The patent dynamically adjusts the warping factor based on the analyzed audio signal characteristics. For speech signals with harmonic structure, the system applies higher warping factors to improve frequency resolution in the critical low-frequency range. For music signals with complex spectral content, the system uses lower warping factors to preserve overall spectral shape and avoid quality deterioration. This dynamic adaptation resolves the contradiction by making the warping characteristic signal-dependent rather than fixed.
Solution Approach 2:
The system changes the warping factor parameter according to signal type analysis. By selecting appropriate warping factor values based on the input signal characteristics, the system achieves high frequency resolution for speech while maintaining audio quality for music. This parameter adaptation allows the same system to optimize for different signal types without compromising either performance.
Data Source
AI summary
An audio encoder, an audio decoder or an audio processor includes a filter for generating a filtered audio signal, the filter having a variable warping characteristic, the characteristic being controllable in response to a time-varying control signal, the control signal indicating a small or no warping characteristic or a comparatively high warping characteristic. Furthermore, a controller is connected for providing the time-varying control signal, which depends on the audio signal. The filtered audio signal can be introduced to an encoding processor having different encoding algorithms, one of which is a coding algorithm adapted to a specific signal pattern. Alternatively, the filter is a post-filter receiving a decoded audio signal.