Echo Suppression Filter Control Information Computation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing acoustic echo suppression systems fail to effectively handle stationary noise in microphone signals, leading to degraded audio quality and performance issues due to systematic estimate deviations, especially in scenarios with low signal-to-noise ratios.

Innovation Solution

The proposed solution involves computing control information for a suppression filter by determining energy-related values for band-pass signals from both audio signals, averaging these values, and modifying them based on their mean values to separate stationary noise from useful signals, thereby improving echo suppression and delay value computation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional echo suppression systems process microphone signals directly, then the system structure remains simple, but the audio quality degrades due to systematic estimate deviations caused by stationary noise

Engineering Contradiction:
Improveecho suppression accuracyVSAvoidsignal processing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the microphone signal into multiple frequency bands using a filter bank, processing each band separately to reduce the impact of stationary noise on overall echo suppression accuracy

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary signal y'[n] that combines the original microphone signal with a noise estimate, serving as a cleaned input for subsequent echo suppression processing

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If the system uses raw energy values from microphone signals, then the computation remains fast, but the accuracy of echo suppression control information degrades due to noise contamination

Engineering Contradiction:
Improvecontrol information accuracyVSAvoidcomputation speed
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent performs preliminary noise estimation and signal cleaning before the main echo suppression computation, preparing cleaner input data that improves subsequent processing accuracy

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent modifies energy-related values by applying corrections based on noise estimates, transforming raw noisy measurements into corrected parameters that accurately reflect the echo component

Inventive Principle:
Principle #35Parameter changes

3Reliability

If the system processes signals without noise separation, then the adaptation speed remains high, but the audio quality suffers from systematic estimate deviations in noisy environments

Engineering Contradiction:
Improveaudio quality consistencyVSAvoidsignal processing steps
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent implements a feedback mechanism where the noise estimate is continuously updated and fed back into the signal processing chain, allowing the system to adapt to changing noise conditions while maintaining consistent audio quality

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The intermediary signal y'[n] acts as a mediator between the noisy microphone input and the echo suppression processor, isolating the main processing from noise contamination

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP2235927B1Apparatus, method and program for computing control information for an echo suppression filter
Publication Date: 2017.07.12 FRAUNHOFER GESELLSCHAFT ZUR FORDERUNG DER ANGEWANDTEN FORSCHUNG EV
  • EP2235927B1 patent drawingFigure 1
  • EP2235927B1 patent drawingFigure 2
  • EP2235927B1 patent drawingFigure 3

AI summary

An embodiment of an apparatus (200) for computing control information for a suppression filter (210) for filtering a second audio signal to suppress an echo based on a first audio signal includes a computation means (220) having a value determination means (230) for determining at least one energy-related value for a band-pass signal of at least two temporally successive data blocks of at least one signal of a group of signals. The computation means (220) further includes a mean value determination means (250) for determining at least one mean value of the at least one determined energy-related value for the band-pass signal. The computation means (220) further includes a modification means (260) for modifying the at least one energy-related value for the band-pass signal on the basis of the determined mean value for the band-pass signal. The computation means (220) ' further includes a control information computation means (270) for computing the control information for the suppression filter (210) on the basis of the at least one modified energy-related value.