Adaptive High-Pass Filter for Speech Coding Noise Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Low bit rate speech coding technologies, such as CELP, often fail to accurately represent short pitch lags in music or singing voice signals, leading to perceptual quality degradation due to double or multiple pitch transmission, resulting in unwanted spectral peaks and discomforting distortion.

Innovation Solution

The implementation of an adaptive high-pass filter with a cut-off frequency below the fundamental frequency is applied to the decoded audio signal when the pitch is less than the minimum allowable pitch, specifically designed to reduce coding noise between 0 and the fundamental frequency, improving perceptual quality by eliminating low-frequency noise.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of substance

If CELP algorithm is used for low bit rate speech coding, then bit rate is reduced and compression is improved, but pitch representation accuracy deteriorates for short pitch lags

Engineering Contradiction:
Improvebit rateVSAvoidpitch representation accuracy
Core Design Contradiction:
Loss of substanceVSMeasurement precision

Solution Approach 1:

The patent converts the harmful effect of coding noise in low-frequency regions into a beneficial outcome by applying adaptive high-pass filtering. The filtering process transforms the degraded signal with unwanted spectral peaks into an improved signal by attenuating noise components below the fundamental frequency, thereby converting the harmful noise into a benefit of enhanced perceptual quality

Inventive Principle:
Principle #22Blessing in disguise (Convert harm into benefit)

Solution Approach 2:

The patent dynamically adjusts the cutoff frequency parameter of the high-pass filter based on the detected fundamental frequency of the speech signal. This parameter adaptation allows the filter to effectively track and suppress coding noise across varying pitch conditions, resolving the contradiction between compression and pitch accuracy

Inventive Principle:
Principle #35Parameter changes

2Reliability

If adaptive high-pass filter is applied to reduce coding noise, then perceptual quality is improved, but device complexity increases

Engineering Contradiction:
Improveperceptual qualityVSAvoidfiltering system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent implements a dynamic adaptive high-pass filter where the cutoff frequency is adjusted in real-time based on the detected fundamental frequency of the speech signal. This dynamic adaptation allows the system to maintain optimal performance across varying speech conditions without requiring a complex fixed-frequency filter bank, thus improving perceptual quality while controlling complexity

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system performs self-adjustment by automatically detecting the fundamental frequency and configuring the high-pass filter parameters accordingly. This self-service mechanism eliminates the need for external manual configuration or complex control systems, achieving improved perceptual quality with minimal additional complexity

Inventive Principle:
Principle #25Self-service

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach enhances the perceptual quality of decoded speech by reducing coding noise and improving the accuracy of pitch representation, resulting in a more natural sound by masking low-frequency noise with harmonic energy, thereby improving the overall quality of music and singing voice signals.

Implementation Method 1

applying an adaptive high pass filter on the decoded audio signal to lower the coding noise at frequencies below the fundamental frequency

Methodology Applied
Scientific EffectHigh-pass filtering: Filter (electronic)

Implementation Method 2

determining a minimum allowable pitch for the CELP algorithm... determining whether the pitch of the audio signal is less than the minimum allowable pitch; and applying an adaptive high pass filter

Methodology Applied
Scientific EffectFrequency selective filtering: Filter (physical)

Data Source

PatentEP2951824B1Adaptive high-pass post-filter
Publication Date: 2020.02.26 HUAWEI TECH CO LTD
  • EP2951824B1 patent drawingFigure 1~2
  • EP2951824B1 patent drawingFigure 3
  • EP2951824B1 patent drawingFigure 4

AI summary

In accordance with an embodiment of the present invention, a method of speech processing included receiving a coded audio signal having coding noise. The method further includes generating a decoded audio signal from the coded audio signal, and determining a pitch corresponding to the fundamental frequency of the audio signal. The method also includes determining the minimum allowable pitch and determining if the pitch of the audio signal is less than the minimum allowable pitch. If the pitch of the audio signal is less than the minimum allowable pitch, applying an adaptive high pass filter on the decoded audio signal to lower the coding noise at frequencies below the fundamental frequency.