Context-Aware Voice Intelligibility Processing in Noisy Playback

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Voice playback devices in noisy environments face challenges in maintaining voice intelligibility due to physical limitations of devices and signal processing constraints, leading to degraded voice quality.

Innovation Solution

The implementation of a voice intelligibility processor that combines digital-to-acoustic level conversion, multiband voice and noise correction, short segment analysis, and global and per-band gain analysis to enhance voice playback, using a system with a microphone, acoustic echo canceler, noise pre-processor, voice intelligibility processor, and loudspeaker, which adjusts voice signals based on device characteristics and noise profiles.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Area of stationary object

If voice playback is performed in noisy environments, then voice playback coverage is improved, but voice intelligibility deteriorates due to background noise masking

Engineering Contradiction:
Improvevoice playback coverageVSAvoidvoice intelligibility
Core Design Contradiction:
Area of stationary objectVSLoss of information

Solution Approach 1:

The patent segments the voice signal into multiple frequency bands and processes each band separately to enhance intelligibility. The noise capture device captures noise across different frequency bands, and the processor applies band-specific gain adjustments to compensate for noise masking effects in each frequency region, thereby maintaining overall voice intelligibility while preserving broad playback coverage.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies local quality enhancement by adjusting gain specifically in frequency bands where noise masking occurs. Rather than uniformly boosting all frequencies, the system identifies problematic frequency regions and applies targeted amplification or spectral shaping to those specific bands, improving intelligibility where needed while maintaining natural sound quality in clean frequency regions.

Inventive Principle:
Principle #3Local quality

2Loss of information

If noise capture device is added to improve voice intelligibility, then voice intelligibility is improved, but device complexity increases

Engineering Contradiction:
Improvevoice intelligibilityVSAvoiddevice complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The patent makes the noise capture device multi-functional by using it for both noise reduction and intelligibility enhancement. The same microphone or audio input that captures ambient noise is also used to analyze the noise spectral profile and guide the intelligibility enhancement process, eliminating the need for separate dedicated sensors and reducing overall device complexity.

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

Solution Approach 2:

The system performs self-service by using its own noise capture capability to automatically adapt to the acoustic environment. The processor continuously monitors the noise spectrum through the existing noise capture device and autonomously adjusts the voice signal processing parameters without requiring external input or manual configuration, thereby maintaining simplicity while achieving adaptive intelligibility enhancement.

Inventive Principle:
Principle #25Self-service

3Loss of information

If signal headroom is increased for voice intelligibility processing, then voice intelligibility is improved, but distortion increases

Engineering Contradiction:
Improvevoice intelligibilityVSAvoiddistortion
Core Design Contradiction:
Loss of informationVSObject-generated harmful factors

Solution Approach 1:

The patent applies dynamic gain adjustment where the processing gain is continuously adapted based on the instantaneous signal level and noise conditions. Rather than applying fixed headroom expansion, the system dynamically modulates the gain in real-time, increasing amplification when the signal is weak and noise is high, and reducing gain when the signal is strong, thereby maintaining intelligibility while avoiding distortion.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system implements feedback control by monitoring the processed output signal and using this information to adjust subsequent processing parameters. The processor analyzes the relationship between input voice signal, noise level, and processed output to automatically regulate the amount of headroom applied, preventing excessive amplification that would cause distortion while ensuring sufficient gain for intelligibility in noisy conditions.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP4029018B1Context-aware voice intelligibility enhancement
Publication Date: 2023.07.26 DTS INC(US)
  • EP4029018B1 patent drawingFigure 1
  • EP4029018B1 patent drawingFigure 2
  • EP4029018B1 patent drawingFigure 3

AI summary

A method comprises: detecting noise in an environment with a microphone to produce a noise signal; receiving a voice signal to be played into the environment through a loudspeaker; performing multiband correction of the noise signal based on a microphone transfer function of the microphone, to produce a corrected noise signal; performing multiband correction of the voice signal based on a loudspeaker transfer function of the loudspeaker to produce a corrected voice signal; and computing multiband voice intelligibility results based on the corrected noise signal and the corrected voice signal.