Audio Signal Separation Modification Parameter Control

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Audio signal separation techniques often fail to fully separate individual sound sources from composite audio signals, leading to degraded audio quality when incomplete separation results in residual components being mixed back into the audio stream, causing undesirable effects like intermediate or dual sound source perceptions.

Innovation Solution

A method determines a separated signal modification parameter based on the success of audio signal separation, limiting the range of modification for characteristics like spatial repositioning, amplitude, and reverberation to prevent unacceptable degradation, and adjusts these parameters dynamically based on the separation success, ensuring the quality of the modified composite audio signal.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If audio signal separation is performed on composite audio signals, then individual sound sources can be identified and separated, but incomplete separation results in residual components being mixed back into the audio stream causing degraded audio quality

Engineering Contradiction:
Improveseparation accuracyVSAvoidaudio quality
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The system calculates a separation success measure by comparing the separated audio signal with the original composite signal and uses this feedback to dynamically adjust the modification parameter. When separation success is high, full modification range is allowed; when separation success is low, modification range is limited to prevent unacceptable degradation of audio quality.

Inventive Principle:
Principle #23Feedback

2Adaptability or versatility

If spatial repositioning and other modifications are applied to separated audio signals, then different audible effects can be provided to listeners, but modifications on incompletely separated signals cause undesirable effects like intermediate or dual sound source perceptions

Engineering Contradiction:
Improveaudible effectsVSAvoidintermediate sound source perception
Core Design Contradiction:
Adaptability or versatilityVSObject-affected harmful factors

Solution Approach 1:

The modification parameter is made dynamic rather than static. It adjusts in real-time based on the calculated separation success measure, allowing the system to adapt the degree of spatial repositioning and other modifications to the current separation quality, thereby preventing harmful effects while maintaining versatility.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes the modification parameter value based on separation success. When separation is successful, full modification range is permitted; when separation is incomplete, the modification parameter is reduced to limit the extent of modifications, preventing intermediate or dual sound source perceptions.

Inventive Principle:
Principle #35Parameter changes

3Adaptability or versatility

If full range modification is applied to separated audio signals regardless of separation success, then maximum audible effects can be achieved, but unacceptable degradation occurs when separation is incomplete

Engineering Contradiction:
Improvemodification rangeVSAvoidaudio signal integrity
Core Design Contradiction:
Adaptability or versatilityVSManufacturing precision

Solution Approach 1:

The system uses feedback from the separation success measure to control the modification parameter. This ensures that full range modification is only applied when separation is successful, maintaining audio signal integrity, while still allowing maximum versatility when conditions permit.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3239981B1Methods, apparatuses and computer programs relating to modification of a characteristic associated with a separated audio signal
Publication Date: 2018.12.12 NOKIA TECHNOLOGIES OY
  • EP3239981B1 patent drawingFigure 1
  • EP3239981B1 patent drawingFigure 2A
  • EP3239981B1 patent drawingFigure 2B~2C

AI summary

This specification describes a method comprising determining, based on a determined measure of success of a separation of an audio signal representing a sound source from a composite audio signal comprising components derived from at least two sound sources, a value of a separated signal modification parameter, the value of the separated signal modification parameter indicating a range of modification of a characteristic associated with the separated audio signal.