Pitch Enhancement Device Using Adaptive Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
When performing pitch enhancement processing on audio signals, long time segments lead to processing delays and inability to accurately enhance pitch components due to fluctuations in the pitch period, resulting in unnatural-sounding decoded audio signals, especially when linear prediction is not involved, and shortening time segments causes discontinuity.
Innovation Solution
Implementing a pitch enhancement apparatus and method that processes audio signals in shorter time segments by using both the current and previous pitch periods to enhance pitch components, with the degree of enhancement decreasing as the time segment becomes more distant, and maintaining energy integrity by considering multiple pitch components.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If pitch enhancement processing is performed on longer time segments, then processing delay is reduced and computational efficiency is improved, but the ability to track pitch period fluctuations deteriorates and discontinuity increases
Solution Approach 1:
The patent applies dynamics by making the pitch enhancement processing adaptive to pitch period fluctuations. The system dynamically adjusts the enhancement based on detected pitch period changes, allowing it to respond to varying pitch conditions while maintaining consistent processing intervals. This resolves the contradiction by enabling accurate pitch tracking without increasing processing delay.
Solution Approach 2:
The patent changes the parameter of pitch period detection and uses this information to modulate the enhancement processing. By detecting pitch period fluctuations and adjusting the enhancement accordingly, the system maintains accuracy despite using fixed-time segment processing, thus resolving the contradiction between processing efficiency and enhancement accuracy.
2Reliability
If pitch enhancement processing is performed on shorter time segments, then pitch period fluctuation tracking is improved and discontinuity is reduced, but processing delay increases and computational efficiency deteriorates
Solution Approach 1:
The patent segments the pitch enhancement processing into distinct stages: pitch period detection, fluctuation determination, and enhancement processing. By segmenting the process, the system can use longer time segments for detection while applying enhancement at appropriate intervals, thus improving accuracy without excessive delay.
Solution Approach 2:
The patent performs preliminary pitch period detection and fluctuation analysis before applying the enhancement processing. This preliminary action allows the system to prepare enhancement parameters in advance, reducing actual processing delay while maintaining accuracy through pre-computed pitch information.
3Measurement precision
If pitch enhancement processing is performed frequently on short time segments, then pitch component enhancement accuracy is improved, but audio signal discontinuity increases
Solution Approach 1:
The patent uses feedback by detecting pitch period fluctuations and using this information to adjust subsequent enhancement processing. The system monitors the audio signal continuously, detects pitch changes, and applies enhancement in a controlled manner that maintains signal continuity while improving pitch accuracy.
Solution Approach 2:
The patent applies dynamics by making the enhancement processing adaptive to detected pitch conditions. The system dynamically adjusts enhancement parameters based on real-time pitch period detection, allowing frequent processing when pitch is stable and reducing processing when fluctuations are detected, thus maintaining continuity while improving accuracy.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A pitch enhancement apparatus obtains an output signal by performing, for each time segment, pitch enhancement processing on a signal derived from an input audio signal. The pitch enhancement apparatus includes a pitch enhancement unit that performs, as the pitch enhancement processing, for each time segment, processing to obtain, as an output signal at each time of the time segment, a signal including the signal at the time, a first signal for enhancement that is the signal at a time which is an earlier time than the time by the number of samples T0 corresponding to a pitch period of the time segment, and a second signal for enhancement that is the signal at a time which is an earlier time than the time by the number of samples T-α corresponding to a pitch period of the α-th time segment previous to the time segment.