Signal Separation via Spatial Rendering for Voice Clarity

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional signal processing systems face a tradeoff between residual noise reduction and distortion in separating desired voice from background noise, and they fail to utilize the human auditory system's ability to distinguish signals localized at different points in space.

Innovation Solution

A signal processing system that uses a multiple rendering section to localize input signals at different positions in space, reducing distortion while maintaining signal separation performance by leveraging the human auditory system's intrinsic separation capabilities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If noise suppression is intensified to reduce residual noise, then the degree of separation between desired voice and background noise is improved, but distortion in the output signal increases significantly

Engineering Contradiction:
Improvedegree of separationVSAvoiddistortion
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The patent segments the monophonic input signal into multiple virtual channel signals through rendering processing, creating separate spatial channels that allow independent manipulation of different signal components while preserving their characteristics

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transforms a monophonic signal into a multi-channel signal by adding spatial dimension through rendering processing, enabling the human auditory organ to utilize its intrinsic ability to separate signals localized at different points in space

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Device complexity

If rendering processing is applied to all signal components uniformly, then the signal processing is simplified, but the ability to separate signals localized at different points in space is lost

Engineering Contradiction:
Improvesignal processing complexityVSAvoidsignal separation capability
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent applies different rendering processing to different input signals based on their characteristics, with each rendering section processing specific signal components differently to achieve optimal separation while maintaining manageable system complexity

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS8892432B2Signal processing system, apparatus and method used on the system, and program thereof
Publication Date: 2014.11.18 NEC CORP

AI summary

Provided is a signal separation system including a rendering unit which receives a first and a second input signal and positions the first input signal according to rendering information.