Audio Processing System for Speech Intelligibility in Noise

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Consuming media content with dialog on mobile devices in noisy environments poses challenges due to ambient noise interference and limitations in audio quality, where existing solutions like noise-cancelling headphones can be costly and disconnect users from important ambient sounds.

Innovation Solution

The development of audio processing systems and methods that create customized audio adjustment profiles for users based on ambient noise conditions, allowing for dialog boost and graphic equalizer settings to be configured and applied, enhancing speech intelligibility without isolating ambient noise.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If noise-cancelling headphones are used to improve speech intelligibility in noisy environments, then speech intelligibility is improved, but the user is disconnected from important ambient sounds and the solution becomes expensive

Engineering Contradiction:
Improvespeech intelligibilityVSAvoidloss of ambient sound awareness
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent extracts only the speech components from the audio signal using speech detection algorithms, rather than blocking all ambient noise. The speech signal is then enhanced and output, allowing users to hear speech clearly while remaining aware of ambient sounds like car horns and sirens.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system uses an intermediary processing stage that detects speech in the audio signal and applies selective enhancement. This intermediary processing allows the system to differentiate between speech and ambient noise, enhancing only the speech portion while preserving ambient sound awareness.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If noise-cancelling headphones are used to improve speech intelligibility, then speech intelligibility is improved, but the solution becomes expensive

Engineering Contradiction:
Improvespeech intelligibilityVSAvoidcost
Core Design Contradiction:
Measurement precisionVSEase of manufacture

Solution Approach 1:

The patent replaces the mechanical/acoustic solution of noise-cancelling headphones with an electronic/digital signal processing solution. By using speech detection algorithms and digital audio processing on mobile devices, the system achieves speech enhancement without requiring expensive specialized hardware.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system leverages the existing multi-functional capabilities of mobile devices (processors, microphones, speakers) to perform speech enhancement, rather than requiring dedicated noise-cancelling hardware. This universal approach uses existing device resources to achieve the desired function at lower cost.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Manufacturing precision

If audio playback is optimized for quiet environments, then audio quality is maintained, but speech intelligibility deteriorates in noisy environments

Engineering Contradiction:
Improveaudio qualityVSAvoidspeech intelligibility in noise
Core Design Contradiction:
Manufacturing precisionVSMeasurement precision

Solution Approach 1:

The system dynamically adapts audio processing based on the ambient environment. Speech detection algorithms continuously analyze the audio signal and adjust enhancement parameters in real-time, allowing the system to optimize speech intelligibility in noisy environments while maintaining natural audio quality in quiet environments.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes audio processing parameters based on detected speech characteristics and ambient noise levels. By dynamically adjusting gain, frequency response, and signal processing parameters, the system maintains audio quality across different environments while improving speech intelligibility when needed.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP4128223B1Systems and methods for enhancing audio in varied environments
Publication Date: 2025.01.01 DOLBY LABORATORIES LICENSING CORP
  • EP4128223B1 patent drawingFigure 1
  • EP4128223B1 patent drawingFigure 2
  • EP4128223B1 patent drawingFigure 3

AI summary

Novel methods and systems for creating and using user profiles for dialog boost and sound equalizer adjustment to compensate for various ambient sound situations. For creating profiles when not at the ambient sound, a synthesized/pre-recorded ambient noise can be mixed with the media to simulate the noise conditions.