Stereo Encoding Circuit Switching Between Time and Frequency Domains
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing low bit rate encoding techniques for speech acoustic signals, such as those using MDCT-based coding, often result in insufficient coding performance due to challenges in handling inter-channel time differences and bit rate limitations.
Innovation Solution
An encoding apparatus and method that determine whether an input stereo signal is suitable for a mid-side stereo scheme, and based on conditions such as the type of signal and inter-channel time difference, either convert the signal into a mid-side signal in the time domain for CELP-based encoding or encode it in the frequency domain using MDCT-based encoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If MDCT-based frequency domain encoding is used for speech signals, then encoding can be performed in frequency domain, but coding performance is insufficient at low bit rates
Solution Approach 1:
The patent changes the encoding parameter by switching between time domain (CELP) and frequency domain (MDCT) encoding methods based on the signal characteristics. When speech signals are detected, the system switches to time domain CELP encoding which provides better coding performance at low bit rates, while using frequency domain MDCT encoding for other signals.
2Device complexity
If mid-side stereo scheme is applied to all signals, then stereo encoding can be simplified, but coding performance deteriorates for speech signals with inter-channel time differences
Solution Approach 1:
The patent applies different encoding qualities to different signal types. For speech signals with inter-channel time differences, the system maintains higher quality by using LR stereo scheme or disabling mid-side conversion, while for other signals where mid-side scheme is appropriate, it applies the simplified mid-side encoding to reduce complexity.
Solution Approach 2:
The patent dynamically adjusts the encoding scheme based on signal characteristics. The system detects whether the input signal is speech and measures inter-channel time differences, then dynamically switches between mid-side and LR stereo schemes or disables mid-side conversion when speech is detected, optimizing performance for varying signal conditions.
3Manufacturing precision
If time domain encoding is used for speech signals, then coding performance improves, but handling inter-channel time differences becomes more difficult
Solution Approach 1:
The patent extracts and removes the problematic mid-side conversion step when speech signals with inter-channel time differences are detected. By disabling the mid-side stereo scheme specifically for speech signals, the system avoids the complexity of handling inter-channel time differences in the mid-side domain while maintaining the benefits of speech-optimized encoding.
Data Source
AI summary
An encoding device includes a control circuit which, when it has been determined that an input stereo signal is suitable to be encoded using a mid-side stereo technique, decides, according to a condition, whether to convert the input stereo signal into a mid-side signal in a time domain and apply first encoding or apply second encoding; a first encoding circuit that encodes the mid-side signal when the first encoding is to be applied; and a second encoding circuit that encodes the input stereo signal in a frequency domain when the second encoding is to be applied.