Audio Encoder Mode Selection Using Estimated Quality Measures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing switched audio coders face complexity challenges in selecting between different encoding algorithms for audio signals, with open-loop mode selection methods performing worse in quality but reducing complexity, while closed-loop methods are more robust but computationally intensive.
Innovation Solution
An apparatus and method for estimating quality measures for each encoding algorithm without actual encoding and decoding, using a controller to select the best algorithm based on these estimates, which includes determining adaptive codebook distortion and considering an innovative codebook to reduce complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If closed-loop mode selection is used to decide between encoding algorithms, then the decision stability and robustness is improved, but the computational complexity increases significantly
Solution Approach 1:
The patent applies preliminary action by computing selection criteria features (such as signal characteristics, spectral properties, and temporal features) before the actual encoding decision is made. This allows the system to pre-evaluate multiple encoding algorithms based on pre-computed features, avoiding the need to perform complete encoding/decoding cycles for each algorithm while still making informed decisions. The selection criteria are prepared in advance based on signal analysis, which stabilizes the decision process without requiring full closed-loop evaluation of each mode.
2Device complexity
If open-loop mode selection is used to reduce computational complexity, then the complexity is reduced significantly, but the decision robustness and quality deteriorate
Solution Approach 1:
The patent implements feedback by using the computed selection criteria to guide the mode selection process. The system calculates features from the audio signal (such as zero-crossing rate, spectral flatness, and energy distribution) and uses these as feedback to determine the appropriate encoding algorithm. This feedback mechanism ensures that the open-loop selection is not arbitrary but is instead guided by quantitative signal characteristics, improving decision robustness while maintaining lower computational complexity compared to full closed-loop approaches.
Solution Approach 2:
The patent applies parameter changes by adjusting the selection criteria thresholds and feature weights based on signal conditions. Instead of using fixed decision rules, the system modifies parameters such as the threshold for switching between coding modes based on the computed signal features. This allows the open-loop selector to adapt to different audio content types (speech, music, transient signals) and make more robust decisions without requiring complete encoding/decoding cycles, thus maintaining quality while reducing complexity.
3Measurement precision
If complete encoding and decoding of both modes is performed to compute selection criteria, then the selection accuracy is improved, but the processing time increases
Solution Approach 1:
The patent extracts only the essential features needed for mode selection from the audio signal, rather than performing complete encoding and decoding. It extracts selection criteria such as spectral properties, temporal characteristics, and energy distribution, which are sufficient to distinguish between speech-like, music-like, and transient signals. By extracting only these critical features and using them to compute selection criteria, the system achieves accurate mode selection without the time penalty of complete encoding/decoding cycles for each algorithm.
Data Source
AI summary
An apparatus for selecting one of a first encoding algorithm having a first characteristic and a second encoding algorithm having a second characteristic for encoding a portion of an audio signal to obtain an encoded version of the portion of the audio signal has a first estimator for estimating a first quality measure for the portion of the audio signal, which is associated with the first encoding algorithm, without actually encoding and decoding the portion of the audio signal using the first encoding algorithm. A second estimator is provided for estimating a second quality measure for the portion of the audio signal, which is associated with the second encoding algorithm, without actually encoding and decoding the portion of the audio signal using the second encoding algorithm. The apparatus has a controller for selecting the first or second encoding algorithms based on a comparison between the first and second quality measures.