Neural Network Audio Signal Reconstruction Circuit

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing audio signal processing technologies fail to effectively reconstruct high-frequency and minute-amplitude components lost due to sampling and quantization with minimal noise and distortion.

Innovation Solution

An audio signal processing device employing a neural network circuit with an input layer, intermediate layers, and an output layer, which simultaneously inputs and processes unit data from consecutive sampling units, utilizing shift registers and multiplier circuits to output reconstructed signals with reduced distortion and noise.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If conventional sampling and quantization methods are used for audio signal processing, then device complexity is reduced and processing efficiency is improved, but high-frequency components and minute-amplitude components are lost resulting in noise and distortion

Engineering Contradiction:
Improveprocessing efficiencyVSAvoidsignal reconstruction accuracy
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The neural network performs preliminary learning offline to establish mapping relationships between sampled signals and original signals. During actual processing, the pre-trained network quickly reconstructs lost components, achieving both efficiency and accuracy.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The neural network acts as an intermediary that bridges the gap between sampled audio data and the original high-fidelity signal, reconstructing lost high-frequency and minute-amplitude components through learned transformations.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Manufacturing precision

If neural network processing is applied to reconstruct lost audio components, then signal reconstruction accuracy is improved, but device complexity and computational load increase

Engineering Contradiction:
Improvesignal reconstruction accuracyVSAvoidneural network circuit complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The neural network weights and structures are predetermined through offline learning, transforming the complex adaptive processing into simpler fixed-pattern processing during runtime, reducing real-time computational burden.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The neural network creates a simplified computational model that copies the essential characteristics of audio signal transformations, enabling complex reconstruction tasks to be performed through simpler predetermined operations.

Inventive Principle:
Principle #26Copying

3Device complexity

If traditional signal processing methods are used, then device complexity is minimized, but noise and distortion in reconstructed signals cannot be effectively reduced

Engineering Contradiction:
Improveprocessing system complexityVSAvoidnoise and distortion
Core Design Contradiction:
Device complexityVSObject-affected harmful factors

Solution Approach 1:

The neural network serves as an intermediary processing layer that specifically targets and reduces noise and distortion components while preserving audio signal quality, achieving harm reduction without requiring complete system redesign.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP3522158B1Audio signal processing device, audio signal processing method, and control program
Publication Date: 2023.08.23 PANASONIC INTELLECTUAL PROPERTY MANAGEMENT CO LTD
  • EP3522158B1 patent drawingFigure 1
  • EP3522158B1 patent drawingFigure 2
  • EP3522158B1 patent drawingFigure 3

AI summary

Provided is an audio signal processing device that is capable of reconstructing, with less noise and distortion, lost components for input signal data generated with a high-frequency component and the like lost due to sampling or the like based on an audio signal string. The audio signal processing device includes a neural network circuit that includes an input layer including input units, an intermediate layer, and an output layer including output units, an input section that executes simultaneous inputting of, at each of unit time intervals, each of pieces of unit data of consecutive sampling units in an input signal data string generated through sampling based on an audio signal string into each of the input units on a one-to-one basis, one of the pieces of unit data input into one of the input units at one of the unit time intervals being input into another of the input units at another of the unit time intervals in the simultaneous inputting at each of the unit time intervals, and an output section that outputs, in accordance with the simultaneous inputting over a plurality of the unit time intervals that are consecutive, a computation result at each of the unit time intervals, the computation result being based on pieces of data output from the output units at each of the unit time intervals.