Location-Based Audio Gain Adjustment for Speech Recognition

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing speech recognition systems in mobile devices face challenges in accurately recognizing speech in environments with varying ambient noise levels, such as factories or warehouses, where noise sources impact performance.

Innovation Solution

A mechanism that adjusts audio system characteristics, including gain and noise models, based on the location of the device, using a network interface to receive location-specific audio properties and apply them to process audio signals for improved recognition accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the mobile device uses a fixed audio processing configuration, then the device complexity is reduced, but the speech recognition accuracy deteriorates in varying noise environments

Engineering Contradiction:
Improvespeech recognition accuracyVSAvoidaudio processing configuration complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent implements dynamic audio processing by adjusting gain and noise model parameters based on detected ambient noise levels. The system transitions from a fixed configuration to a dynamic one where processing characteristics change in real-time according to environmental conditions, thereby maintaining speech recognition accuracy across varying noise environments without requiring multiple fixed configurations.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes audio processing parameters (gain values, noise model characteristics) based on the detected noise environment. By modifying these parameters dynamically rather than maintaining a fixed configuration, the system achieves adaptability to different acoustic conditions while managing complexity through parameter adjustment rather than structural changes.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If the device adjusts audio processing parameters dynamically based on location, then speech recognition accuracy in varying noise environments is improved, but the computational requirements and processing time increase

Engineering Contradiction:
Improvespeech recognition accuracyVSAvoidaudio processing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary actions by pre-establishing noise models for different locations and preparing gain adjustment strategies in advance. When the device detects its location or ambient noise level, it can quickly apply pre-computed processing parameters rather than performing complex real-time analysis, thereby reducing processing time while maintaining accuracy.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system uses feedback from ambient noise detection to adjust audio processing parameters. By continuously monitoring the noise environment and applying feedback-based adjustments to gain and noise model parameters, the system achieves adaptive speech recognition with reduced processing time compared to exhaustive real-time optimization methods.

Inventive Principle:
Principle #23Feedback

3Object-affected harmful factors

If the device uses location-specific noise models, then the ability to compensate for ambient noise is improved, but the data storage requirements increase

Engineering Contradiction:
Improveambient noise interferenceVSAvoidstored noise model data
Core Design Contradiction:
Object-affected harmful factorsVSQuantity of substance

Solution Approach 1:

The patent applies local quality by creating location-specific noise models that capture the acoustic characteristics of particular environments. Each location has its own tailored noise model rather than using a generic one, enabling precise compensation for local ambient noise conditions. This approach optimizes noise compensation effectiveness while managing storage by focusing on location-specific rather than universal models.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP3309780B1Utilization of location and environment to improve recognition
Publication Date: 2021.09.15 VOCOLLECT INC
  • EP3309780B1 patent drawingFigure 1~2
  • EP3309780B1 patent drawingFigure 3
  • EP3309780B1 patent drawingFigure 4~5

AI summary

A portable terminal has a network interface that receives a set of instructions having a sequence of at least one location and audio properties associated with the at least one location from a server. An audio circuit receives audio signals picked up by a microphone and processes the audio signals in a manner defined by the audio properties associated with the at least one location. A speech recognition module receives processed signals from the audio circuit and carries out a speech recognition process thereupon.