Audio Signal Recovery via FFT Brick Wall Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The defects caused by filtered high-frequency signals in audio data, which become more apparent with advancements in digital-to-analog converters and earphones, necessitate a method for recovering audio signals to improve perceived audio quality.

Innovation Solution

A method involving a terminal with a processor, memory, and transceiver that buffers audio signals, performs FFT, identifies and recovers frequency subbands, and separates and synthesizes high-frequency signals using MDCT and IIR filtering to restore the audio signal.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of energy

If low-pass filtering is applied to audio data to filter high-frequency signals, then transmission resources are saved and compression ratio is increased, but audio quality deteriorates due to loss of high-frequency components

Engineering Contradiction:
Improvetransmission resourcesVSAvoidaudio quality
Core Design Contradiction:
Loss of energyVSManufacturing precision

Solution Approach 1:

The patent performs preliminary analysis of the audio signal to detect the brick wall frequency and identify lost high-frequency components before reconstruction. By detecting the cutoff frequency characteristics in advance and preparing reconstruction filters accordingly, the system can effectively recover high-frequency signals while maintaining efficient transmission of the original low-pass filtered audio data.

Inventive Principle:
Principle #10Preliminary action

2Quantity of substance

If high-frequency signals are filtered out to reduce audio data amount, then data transmission efficiency is improved, but perceived audio quality becomes obviously defective

Engineering Contradiction:
Improveaudio data amountVSAvoidperceived audio quality
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

The patent introduces an intermediary reconstruction process that analyzes the compressed audio signal to detect the brick wall frequency, identifies the range of lost high-frequency components, and synthesizes reconstruction signals to fill in the missing frequency spectrum. This intermediary analysis and synthesis step bridges the gap between the compressed audio data and the perceived audio quality requirement.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Productivity

If audio data is compressed to reduce data amount, then transmission resources are saved, but coding artifacts are introduced that degrade audio quality

Engineering Contradiction:
Improvecompression ratioVSAvoidaudio quality
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The patent converts the harmful effect of brick wall frequency artifacts and missing high-frequency components into a beneficial reconstruction process. By detecting the characteristic brick wall frequency pattern introduced by compression and using it as a guide to identify exactly which frequency components need reconstruction, the system transforms the compression artifacts into information that guides the recovery process, effectively turning the harm of data reduction into an opportunity for targeted signal restoration.

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

Data Source

PatentEP3644312B1Method and apparatus for recovering audio signals
Publication Date: 2023.10.11 GUANGZHOU KUGOU COMP TECH CO LTD
  • EP3644312B1 patent drawingFigure 1
  • EP3644312B1 patent drawingFigure 2
  • EP3644312B1 patent drawingFigure 3

AI summary

The present disclosure, belonging to the field of audio technology, provides a method and apparatus for recovering audio signals. The method includes: buffering an audio signal sampled at a preset number of sampling points each time, and then performing frequency spectrum analysis on the sampled audio signal by FFT; when it is determined that the audio signal is compressed, filtering a frequency point; recovering high-frequency signals based on audio signals before the frequency point; and performing phase recovery on the high-frequency signals. The present disclosure provides the method for recovering audio signals.