Controlling voice recognition sensitivity for voice recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern washing machines with speech recognition capabilities face challenges in maintaining optimal speech recognition sensitivity, leading to misrecognition of user inputs due to varying speech volumes and surrounding noise, without adequate adjustment of sensitivity settings.
Innovation Solution
A device with a processor and memory that adjusts speech recognition sensitivity based on pre-trained models and user-specific settings, optimizing the sensitivity to enhance recognition success rates and minimize misrecognition by varying confirmation sound volumes according to recognition scores.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If speech recognition sensitivity is increased to recognize speech in noisy environments, then speech recognition capability is improved, but misrecognition of surrounding noise as speech increases
Solution Approach 1:
The patent implements dynamic adjustment of speech recognition sensitivity based on real-time environmental noise levels. The system continuously monitors the acoustic environment and adapts the recognition threshold accordingly, transitioning from a static sensitivity setting to a dynamic one that responds to changing conditions. This resolves the contradiction by allowing high sensitivity in noisy environments while maintaining accuracy through adaptive threshold adjustment.
Solution Approach 2:
The system changes the speech recognition sensitivity parameter based on detected noise levels and user speech characteristics. By adjusting the recognition threshold parameter dynamically, the system optimizes the balance between capturing speech in noisy conditions and avoiding false recognition of background noise. This parameter adaptation directly addresses the trade-off between recognition capability and accuracy.
2Measurement precision
If speech recognition sensitivity is decreased to avoid misrecognition, then speech recognition accuracy is improved, but speech recognition capability in noisy environments deteriorates
Solution Approach 1:
The system dynamically adjusts sensitivity based on environmental conditions rather than using a fixed low threshold. When noise levels are high, the system temporarily increases sensitivity to maintain recognition capability, while using advanced signal processing to preserve accuracy. This dynamic approach resolves the contradiction by allowing context-dependent sensitivity adjustment.
Solution Approach 2:
The patent introduces noise filtering and speech detection algorithms as intermediary processes between the microphone input and speech recognition engine. These intermediaries preprocess the audio signal to enhance speech components and suppress background noise, allowing the system to maintain both high sensitivity and high accuracy simultaneously by improving the quality of input signals.
3Device complexity
If constant speech recognition sensitivity is used, then device complexity is reduced, but adaptability to varying speech volumes and noise levels deteriorates
Solution Approach 1:
The system performs self-adjustment of speech recognition sensitivity without requiring manual user configuration. It automatically monitors environmental noise levels and user speech patterns, then adapts the recognition parameters autonomously. This self-service capability provides high adaptability while keeping the user interface simple, effectively resolving the contradiction between complexity and adaptability.
Solution Approach 2:
The patent implements feedback mechanisms where the system monitors recognition outcomes and environmental conditions, then uses this information to adjust sensitivity settings. The feedback loop continuously optimizes performance by comparing expected speech patterns with actual inputs and adapting parameters accordingly, providing adaptability through a relatively simple feedback-based control mechanism.
Data Source
AI summary
A device for changing a speech recognition sensitivity for speech recognition can include a memory and a processor configured to obtain a first plurality of speech data input at different times, apply a pre-trained speech recognition model to the first plurality of speech data at a plurality of different speech recognition sensitivities, obtain a first speech recognition sensitivity from among the plurality of different speech recognition sensitivities based on the pre-trained speech recognition model and the plurality of different speech recognition sensitivities, the first speech recognition sensitivity corresponding to an optimal speech recognition sensitivity at which a speech recognition success rate of the speech recognition model satisfies a set first recognition success rate criterion, and change a setting of the speech recognition sensitivity based on the first speech recognition sensitivity obtained from among the plurality of different speech recognition sensitivities.