Noise Estimation With Music Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing audio signal processing systems face unpredictable performance and decreased signal quality when loud music with speech-like components is present, as they fail to accurately model music content and adapt noise estimates accordingly.

Innovation Solution

A system and method for noise estimation with music detection that classifies audio signals as music or non-music, adjusting the adaptation rate of noise estimation to mitigate the impact of music components, thereby providing a conservative noise estimate and improving signal quality in noisy environments with both speech and music.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If noise modeling methods assume background noise does not contain speech-like content, then noise estimation works for typical environmental noise, but the system acts unpredictably and decreases signal quality when music with speech-like components is present

Engineering Contradiction:
Improvenoise estimation reliabilityVSAvoidadaptability to music content
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The system changes the adaptation rate parameter based on music detection. When music is detected, the adaptation rate is reduced to prevent the noise estimator from incorrectly adapting to music content. When no music is present, the adaptation rate returns to normal levels, allowing effective noise estimation for environmental sounds.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

A music detector acts as an intermediary component between the audio input and the noise estimator. The music detector analyzes the audio signal to determine whether music content is present, and based on this analysis, it controls the adaptation rate of the noise estimator, thereby mediating the interaction between music content and noise estimation process.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If the noise estimator continuously adapts to environmental noise, then noise modeling accuracy improves over time, but the presence of music causes the noise estimate to be misled and act unpredictably

Engineering Contradiction:
Improvenoise estimation accuracyVSAvoidsignal quality stability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The system dynamically adjusts the adaptation rate based on real-time music detection. The adaptation rate is not fixed but changes over time according to the presence or absence of music content. This dynamic control allows the system to maintain high noise estimation accuracy when appropriate while preventing unpredictable behavior when music is present.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The music detector performs preliminary analysis of the audio signal to detect music content before the noise estimator can be misled. By detecting music in advance, the system can preemptively reduce the adaptation rate, preventing the noise estimate from being incorrectly influenced by music content with speech-like components.

Inventive Principle:
Principle #9Preliminary anti-action

3Stability of the object's composition

If the system reduces adaptation rate in presence of music, then noise estimation stability improves, but the ability to quickly adapt to actual noise changes may be reduced

Engineering Contradiction:
Improvenoise estimate stabilityVSAvoidadaptation speed
Core Design Contradiction:
Stability of the object's compositionVSSpeed

Solution Approach 1:

The adaptation rate is dynamically controlled based on music detection results. When music is detected, the adaptation rate is reduced to maintain stability. When music is absent, the adaptation rate increases to allow quick adaptation to actual noise changes. This dynamic adjustment resolves the contradiction by making the adaptation speed conditional on the audio content type.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP3349213B1System and method for noise estimation with music detection
Publication Date: 2020.07.01 BLACKBERRY LTD
  • EP3349213B1 patent drawingFigure 1
  • EP3349213B1 patent drawingFigure 2
  • EP3349213B1 patent drawingFigure 3

AI summary

In a system and method for noise estimation with music detection described herein provides for generating a music classification for music content in an audio signal. The music detector may classify the audio signal as music or non-music. The non-music signal may be considered to be signal and noise. An adaption rate may be adjusted responsive to the generated music classification. A noise estimate is calculated applying the adjusted adaption rate. The system and method may mitigate the noise modeling algorithms being misled by the music components.