Source-Separated Virtual Bass Enhancement Without Distortion

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing virtual bass enhancement algorithms suffer from intermodulation distortion and computational complexity, particularly in time-domain techniques, while frequency-domain techniques suffer from smearing effects, both of which impair the perception of low-frequency sounds in small loudspeakers.

Innovation Solution

The method involves applying virtual bass enhancement (VBE) algorithms selectively to isolated music stems or acoustic sources using a music demixing model, such as neural networks, to extract audio channels, and then processing these channels independently with time-domain or frequency-domain techniques, followed by post-processing stages to generate overtones and combine them, thereby avoiding distortion and maintaining low computational complexity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If time-domain VBE algorithms are applied to enhance bass perception, then bass enhancement is achieved, but intermodulation distortion is introduced

Engineering Contradiction:

Improvebass enhancementVSAvoidintermodulation distortion

Core Design Contradiction:

ReliabilityVSObject-generated harmful factors

Solution Approach 1:

The patent applies segmentation by separating the audio signal into different frequency bands (low-frequency components below cutoff and high-frequency components above cutoff) and processing them through different pathways. The low-frequency components are processed through nonlinear devices to generate virtual bass, while high-frequency components are processed separately and then combined, preventing intermodulation distortion between different frequency components.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent extracts the harmful intermodulation distortion by separating the nonlinear processing from the full-band signal. Instead of applying nonlinear devices to the complete audio signal, the method extracts only the low-frequency components below the cutoff frequency for nonlinear processing, thereby taking out the source of intermodulation distortion while preserving the bass enhancement effect.

Inventive Principle:

Principle #2Taking out (Extraction)

2Reliability

If hybrid VBE techniques are applied to merge advantages of time-domain and frequency-domain methods, then performance is improved, but computational cost increases

Engineering Contradiction:

Improvebass enhancement performanceVSAvoidcomputational cost

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the audio signal processing into distinct frequency-based pathways, applying simple nonlinear processing only to low-frequency components below the cutoff frequency. This segmentation avoids the need for complex hybrid algorithms that would process the entire spectrum with high computational cost, while still achieving effective bass enhancement through the nonlinear generation of harmonics from the low-frequency portion.

Inventive Principle:

Principle #1Segmentation

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 effectively enhances bass perception without introducing distortion and reduces computational demands, providing improved bass reproduction in small loudspeakers by allowing tailored processing for each acoustic source, thus overcoming the limitations of existing VBE algorithms.

Implementation Method 1

a music demixing model, such as neural networks, to extract audio channels

Methodology Applied

Scientific EffectSource separation:

Implementation Method 2

at least one virtual bass enhancing unit configured to generate overtones for enhancing a bass perception

Methodology Applied

Scientific EffectOvertone generation:

Implementation Method 3

at least one adder configured to add the overtones to the input audio signal

Methodology Applied

Scientific EffectSignal summation:

Data Source

PatentUS12593194B2Virtual bass enhancement based on source separation

Publication Date: 2026.03.31 INVENTVM SEMICON SRL

AI summary

A virtual bass enhancing device for enhancing a virtual bass of an input audio signal includes a demixer, configured to extract at least one audio channel from the input audio signal, wherein the audio channel corresponds to an acoustic source, or to a group of acoustic sources, of the input audio signal, at least one virtual bass enhancing unit configured to generate overtones for enhancing a bass perception of the audio channel, and at least one adder configured to add the overtones to the input audio signal so as to generate an enhanced audio signal.