Mobile Speech Noise Suppression Using Accelerometer Correlation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing non-stationary noise suppression processes in mobile communication devices often mistakenly treat speech signals as ambient noise, leading to unwanted muting, and fail to effectively differentiate between speech and ambient noise.

Innovation Solution

A mobile communication device equipped with a primary microphone, an accelerometer, and a processor that generates a noise-suppressed speech signal by cross-correlating the primary speech signal with a signal from the accelerometer, identifying and suppressing regions of strong cross-correlation, and reducing amplitude in those regions to separate speech from ambient noise.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Object-affected harmful factors

If known non-stationary noise suppression processes are used, then ambient noise is suppressed, but speech signal is mistakenly treated as noise and muted

Engineering Contradiction:

Improveambient noise suppressionVSAvoidspeech signal integrity

Core Design Contradiction:

Object-affected harmful factorsVSReliability

Solution Approach 1:

The patent introduces a secondary microphone as an intermediary element to capture ambient noise separately from the primary speech signal. This mediator allows the system to distinguish between speech and noise by comparing signals from two microphones, preventing the speech signal from being mistakenly suppressed while still achieving noise reduction through differential processing.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Object-affected harmful factors

If primary and secondary microphones are positioned at predetermined distance, then noise dampening is achieved, but speech signal level difference causes suppression errors

Engineering Contradiction:

Improvenoise dampeningVSAvoidspeech level differentiation

Core Design Contradiction:

Object-affected harmful factorsVSMeasurement precision

Solution Approach 1:

The patent applies dynamic gain adjustment based on the calculated speech signal level difference between microphones. Instead of using fixed positioning and static processing, the system dynamically adapts the suppression parameters according to real-time speech level detection, allowing accurate differentiation between speech and noise even when microphones are positioned at predetermined distances for noise dampening.

Inventive Principle:

Principle #15Dynamics

3Object-affected harmful factors

If speech signal level at primary microphone is significantly louder than at secondary microphone, then noise suppression is effective, but speech muting occurs when levels are equal

Engineering Contradiction:

Improvenoise suppression effectivenessVSAvoidsuppression accuracy across different speech conditions

Core Design Contradiction:

Object-affected harmful factorsVSAdaptability or versatility

Solution Approach 1:

The patent changes the processing parameters dynamically based on the speech signal level difference detected between microphones. When speech levels are equal (indicating poor positioning or background speech), the system adjusts suppression parameters to prevent muting. When speech levels show clear differentiation, normal suppression effectiveness is maintained. This parameter adaptation resolves the contradiction between suppression effectiveness and accuracy across varying speech conditions.

Inventive Principle:

Principle #35Parameter changes

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Effectively suppresses ambient noise without muting the speech signal, ensuring clear communication by accurately distinguishing between speech and ambient noise through advanced signal processing techniques.

Implementation Method 1

an accelerometer configured to detect vibrations of the mobile communication device when the speech and the ambient noise are received at the microphone and to generate a signal corresponding to the vibrations that are detected

Methodology Applied

Scientific EffectVibration detection: Vibration

Implementation Method 2

generating, at the processor, using signal processing, a noise suppressed speech signal based on the primary speech signal and the signal received from the accelerometer

Methodology Applied

Scientific EffectCross-correlation:

Implementation Method 3

generating, at the processor, using signal processing, a noise suppressed speech signal based on the primary speech signal and the signal received from the accelerometer

Methodology Applied

Scientific EffectAmplitude reduction:

Data Source

PatentEP3163572B1Method and device for supressing ambient noise in a speech signal generated at a microphone of the device

Publication Date: 2019.09.18 BLACKBERRY LTD

Figure 1
Figure 2
Figure 3

AI summary

The present disclosure relates to a mobile communication device and a method of controlling a mobile communication device to suppress ambient noise in a speech signal generated at a microphone of the mobile communication device. A primary microphone configured to receive speech and ambient noise and generate a primary speech signal corresponding to the speech and the ambient noise that is received. An accelerometer configured to detect vibrations of the mobile communication device when the speech and the ambient noise are received at the microphone and to generate a signal corresponding to the vibrations that are detected. A processor is operably coupled to the microphone and the accelerometer and configured to generate, using signal processing, a noise suppressed speech signal based on the primary speech signal and the signal received from the accelerometer.