Speech Signal Post-Processing Using Pitch Correction and Enhancement Filters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current post-processing techniques for speech signals in low bit-rate speech coding suffer from high computational complexity and auditory distortions due to the use of multiple bandpass filters and excessive multiplications, which are inefficient for pitch enhancement.
Innovation Solution
The proposed method employs a sequence of pitch correction, weight adjustment, and enhancement filters to process speech signals, reducing complexity by using a combination of pitch correction filter, pitch weight parameter adjustor, and pitch enhancement filters, which can be ordered in various configurations to achieve simultaneous pitch emphasis and enhancement with lower computational demands.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple bandpass filters are used to enhance pitch components in certain frequency bands, then pitch enhancement effect is improved, but computational burden increases significantly
Solution Approach 1:
The patent combines multiple bandpass filters into a single filter structure that processes all frequency bands simultaneously. Instead of separately filtering and enhancing each frequency band, the invention merges these operations into one unified filter that achieves both bandpass filtering and pitch enhancement, thereby reducing computational complexity while maintaining effectiveness.
Solution Approach 2:
The single filter structure in the patent serves multiple functions simultaneously: it acts as a bandpass filter, a pitch enhancement filter, and a computational efficiency optimizer. This multi-functional design eliminates the need for separate processing stages for each frequency band, reducing overall computational burden while achieving comprehensive pitch enhancement.
2Reliability
If adaptive codebook driven excitation is directly added into total excitation, then pitch enhancement is achieved, but computational complexity increases due to multiplications and square computations
Solution Approach 1:
The patent extracts and removes the computationally intensive multiplication and square root operations from the pitch enhancement process. By eliminating these complex computations and replacing them with simpler filter-based operations, the invention maintains pitch enhancement effectiveness while significantly reducing computational complexity.
Solution Approach 2:
The patent replaces the mechanical computation-based approach (multiplications and square roots) with a signal processing-based approach using filter operations. This substitution transforms the computational process from arithmetic-heavy operations to convolution-based filtering, which are more efficient and less computationally demanding.
3Reliability
If traditional post-processing techniques are used in low bit-rate speech coding, then pitch distortion is reduced, but auditory distortions and computational complexity increase
Solution Approach 1:
The patent applies local quality enhancement by targeting specific frequency bands with tailored filter characteristics. Instead of uniformly processing all frequency components, the invention designs filters that selectively enhance pitch components in specific bands while preserving natural speech characteristics in other bands, thereby reducing auditory distortions while maintaining pitch accuracy.
Data Source
AI summary
A method for post-processing of speech signals includes using a pitch correction filter, a pitch weight parameter adjustor, and a first pitch enhancement filter to process the input signal into a first output signal; summing both the input signal and the first output signal as a second output signal; and using a second pitch enhancement filter to process the second output signal. Furthermore, another method for post-processing of speech signals includes using a second pitch enhancement filter to process the input signal into a second output signal; using a pitch correction filter, a pitch weight parameter adjustor, and a first pitch enhancement filter to process the second output signal into a first output signal; and summing both the second output signal and the first output signal as a final output signal. The two methods can simultaneously realize pitch emphasis and enhancement with low computation complexity.