Embedded Sound Signal Coding Reducing Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing monaural coding schemes require high algorithm latency and significant arithmetic processing to obtain monaural local decoded signals, which is problematic for applications requiring low latency and minimal processing overhead, especially in embedded coding/decoding of 2-channel sound signals.
Innovation Solution
A sound signal coding method that involves obtaining a downmix signal by mixing left and right channel input sound signals, calculating and coding subtraction gains, and using these gains to generate monaural and stereo codes, thereby reducing latency and processing requirements while maintaining sound quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a high-quality monaural coding scheme (e.g., 3GPP EVS standard) is used to obtain monaural local decoded signals, then sound quality of decoded sound signals is improved, but algorithm latency increases beyond frame length
Solution Approach 1:
The patent extracts only the necessary monaural coding functionality without requiring the full monaural local decoded signals. Instead of obtaining complete decoded signals through complex decoding processing, the invention uses a simplified approach that extracts essential monaural information directly from the input sound signals, thereby avoiding the latency issue while maintaining sound quality improvement benefits
Solution Approach 2:
The patent performs preliminary monaural coding processing on the input sound signals before the main stereo coding process. By pre-processing the sound signals to obtain monaural components and their spectral characteristics in advance, the system can improve sound quality without requiring post-decoding operations that would introduce latency
2Measurement precision
If monaural local decoded signals are obtained through decoding processing in the coding device, then sound quality is improved, but arithmetic processing amount increases
Solution Approach 1:
The patent extracts essential monaural information (spectral envelope, power spectrum) directly from the input sound signals without performing complete decoding processing. This extraction approach obtains the necessary monaural characteristics with significantly reduced computational effort compared to full decoding operations
Solution Approach 2:
Instead of performing complex decoding processing to obtain monaural local decoded signals, the patent creates simplified copies or representations of monaural information (such as spectral envelopes and power spectra) that capture the essential characteristics needed for sound quality improvement but require far less arithmetic processing
Data Source
AI summary
A downmix unit 110 obtains downmix signals which are signals obtained by mixing input sound signals of a left channel input and input sound signals of a right channel input. A left channel signal subtraction unit 130 and a right channel signal subtraction unit 150 code the difference between the input sound signals and a multiplication value of the downmix signals and a subtraction gain for each of the left channel and the right channel. In such a configuration, a left channel subtraction gain estimation unit 120 and a right channel subtraction gain estimation unit 140 determine the subtraction gain such that the quantization errors resulting from the two processes of coding/decoding are reduced.