Audio Signal Processing Stage for Small Loudspeaker Bass Reproduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Small loudspeakers struggle to faithfully reproduce low-frequency audio content due to limited membrane excursion, leading to potential distortions and inefficiencies, with existing solutions either over-attenuating low frequencies or causing perceivable loss of content.

Innovation Solution

An audio signal processing stage comprising a filter bank, compressor branches, a virtual bass unit, and a summation unit to separate and process audio signals into frequency bands, compressing signals to prevent overdrive while generating harmonics to enhance low-frequency content, thereby reducing distortion and improving audio quality.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If frequency attenuation techniques are used to prevent loudspeaker saturation at low frequencies, then distortion is reduced, but low frequency content is excessively attenuated

Engineering Contradiction:
Improvedistortion preventionVSAvoidlow frequency content
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

The audio signal is divided into multiple frequency bands using a filter bank, allowing independent processing of each band. This segmentation enables targeted compression of specific frequency ranges without affecting others, preventing excessive attenuation of low frequency content while still preventing distortion.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system uses dynamic compression with adjustable parameters for each frequency band, allowing the attenuation level to adapt based on the input signal characteristics. This dynamic approach prevents fixed excessive attenuation while maintaining distortion prevention, preserving more low frequency content compared to static methods.

Inventive Principle:
Principle #15Dynamics

2Reliability

If compression is applied to prevent loudspeaker overdrive, then distortion is reduced, but device complexity increases

Engineering Contradiction:
Improveoverdrive preventionVSAvoidsignal processing complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The signal processing is segmented into multiple independent frequency band processors. Each band can be processed with simple compression algorithms, and the results are combined. This segmentation allows complex overall processing to be achieved through multiple simple, manageable stages, reducing the complexity burden on any single component.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The filter bank and compression architecture provides multi-functionality by handling multiple frequency bands simultaneously with a unified processing framework. This universal structure can process different frequency ranges using the same basic compression mechanism, reducing overall system complexity compared to having separate dedicated processors for each band.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Reliability

If existing adaptive equalization methods are used, then loudspeaker saturation is prevented, but perceivable loss of low frequency content occurs

Engineering Contradiction:
Improvesaturation preventionVSAvoidlow frequency content
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

Different processing characteristics are applied to different frequency bands. Low frequency bands receive compression with parameters optimized for preserving bass content, while other bands receive appropriate processing for their characteristics. This local quality approach ensures saturation prevention without uniform excessive attenuation of low frequencies.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The compression parameters (threshold, ratio, attack, release) are specifically optimized for each frequency band, particularly for low frequencies. By changing parameters locally for different bands rather than using uniform settings, the system prevents saturation while preserving perceivable low frequency content that would be lost with generic adaptive equalization.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3453187B1Audio signal processing stage, audio signal processing apparatus and audio signal processing method
Publication Date: 2020.05.13 HUAWEI TECH CO LTD
  • EP3453187B1 patent drawingFigure 1
  • EP3453187B1 patent drawingFigure 2
  • EP3453187B1 patent drawingFigure 3

AI summary

The invention relates to an audio signal processing stage (600) for processing an input audio signal (101) into an output audio signal (103). The audio signal processing stage (600) comprises: a filter bank (105a) defining two or more frequency bands, the filter bank separating the input audio signal (101) into two or more input audio signal components (X(k,b)); a set of two or more band branches (105e) which provide two or more output audio signal components (Y(k,b)). The set of two or more band branches (105e) comprises one or more compressor branches, each comprising a compressor (105b) which compresses the respective input audio signal component (X(k,b)) to provide the respective output audio signal component (Y(k,b)); an inverse filter bank (105d) which generates a summed audio signal (y(t)) by summing the two or more output audio signal components (Y(k,b)); a residual audio signal generating unit (613) which generates a residual audio signal (v(t)) which is a difference between the input audio signal (101) and the summed audio signal(y(t)); a virtual bass unit (107) which generates a virtual bass signal (w(t)) that comprises one or more harmonics of the residual audio signal (v(t)); and a summation unit (109) which generates the output audio signal (103) by summing the summed audio signal (y(t)) and the virtual bass signal (w(t)). A variant (700) of the audio signal processing stage (600), and corresponding audio signal processing methods are also described.