Spectrum Coding Apparatus Maintaining Harmonic Structure
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional coding technologies distort the harmonic structure of voice and audio signals when compressing them to low bit rates, leading to a deterioration in sound quality due to the lack of consideration for the spectral shape during the replacement of high-frequency domains by low-frequency domains.
Innovation Solution
A coding apparatus and method that acquires spectra divided into low-frequency and high-frequency bands, calculates a parameter indicating the degree of similarity between the two based on the harmonic structure, and encodes this parameter instead of the high-frequency spectrum, ensuring that the harmonic structure is maintained during coding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If the spectrum of the high-frequency domain is replaced by the spectrum of the low-frequency domain to reduce bit rate, then the bit rate is reduced, but the harmonic structure is collapsed and sound quality deteriorates
Solution Approach 1:
The patent changes the parameter representation from direct spectrum values to parameters that define harmonic structure (fundamental frequency, amplitude, phase). By encoding these parameters instead of raw spectral data, the patent maintains the harmonic structure while reducing bit rate, as fewer parameters are needed to represent the periodic pattern than to represent the full spectrum.
Solution Approach 2:
The patent segments the spectrum representation into distinct harmonic components (fundamental frequency and its multiples) rather than treating it as a continuous spectrum. This segmentation allows selective encoding of only the essential harmonic parameters, maintaining structural integrity while reducing data quantity.
2Loss of energy
If narrow band signals (300 Hz to 3.4 kHz) are used for coding to achieve high quality at low bit rates, then bit rate is reduced, but the frequency band is limited and sound lacks realism
Solution Approach 1:
The patent dynamically adapts the frequency band utilization by detecting and encoding harmonic structures that can extend into higher frequency ranges. Instead of being constrained to a fixed narrow band, the system dynamically extends the effective frequency coverage by exploiting the periodic harmonic patterns that naturally occur in speech signals, allowing wider frequency representation without proportionally increasing bit rate.
3Area of stationary object
If wideband signals (50 Hz to 7 kHz) are used for coding to expand frequency band, then frequency band is expanded, but bit rate increases
Solution Approach 1:
The patent transforms the coding approach from representing all wideband spectral data to representing only the essential harmonic parameters (fundamental frequency, amplitude, phase). This parameter transformation allows the system to cover a wide frequency band by mathematically generating harmonic multiples, significantly reducing the bit rate required compared to encoding the full wideband spectrum directly.
Data Source
AI summary
A coding apparatus capable of coding a spectrum at a low bit rate and with high quality without producing any disturbance in a harmonic structure of the spectrum. In this apparatus, internal state setting section (106) sets an internal state of filtering section (107) using first spectrum S1(k). Pitch coefficient setting section (109) outputs pitch coefficient T by gradually changing it. Filtering section (107) calculates estimated value S′2(k) of second spectrum S2(k) based on pitch coefficient T. Search section (108) calculates the degree of similarity between S2(k) and S′2(k). At this time, pitch coefficient T′ corresponding to the maximum calculated degree of similarity is given to filter coefficient calculation section (110). Filter coefficient calculation section (110) determines filter coefficient βi using this pitch coefficient T′.