Speech Recognition Device Distance-Based Expression Selection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing speech recognition systems face increased rates of false recognition due to background noise, which current techniques struggle to mitigate effectively.

Innovation Solution

A speech recognition device and method that determines the distance to a speech source and adjusts the set of recognizable registered expressions based on this distance, using a recognition target table to vary the number and type of expressions that can be recognized, thereby reducing the impact of background noise.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multiple standard patterns are used for speech recognition, then recognition accuracy improves, but false recognition rate increases in noisy environments

Engineering Contradiction:

Improvespeech recognition accuracyVSAvoidfalse recognition rate

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The patent dynamically adjusts the set of recognizable registered expressions based on the determined distance to the speech source. As distance increases, the system reduces the number of recognizable expressions to those most likely to be uttered at that distance, making the recognition system adaptive to changing acoustic conditions rather than static

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes the parameter of recognizable expressions based on distance measurements. By modifying which expressions are recognized as valid targets based on the distance parameter, the system optimizes recognition accuracy for the current acoustic environment while reducing false positives from background noise

Inventive Principle:

Principle #35Parameter changes

2Adaptability or versatility

If the number of recognizable expressions is increased, then speech recognition coverage improves, but false recognition due to background noise increases

Engineering Contradiction:

Improverecognition coverageVSAvoidbackground noise impact

Core Design Contradiction:

Adaptability or versatilityVSObject-affected harmful factors

Solution Approach 1:

The system dynamically adjusts the set of recognizable registered expressions based on the determined distance to the speech source. As distance increases, the system reduces the number of recognizable expressions to those most likely to be uttered at that distance, making the recognition system adaptive to changing acoustic conditions rather than static

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

Different sets of expressions are recognized based on local acoustic conditions (distance). The system applies different recognition criteria locally - recognizing more expressions when the speech source is close and fewer expressions when it is far - rather than using a uniform recognition set throughout

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS10540972B2Speech recognition device, speech recognition method, non-transitory recording medium, and robot

Publication Date: 2020.01.21 CASIO COMPUTER CO LTD

AI summary

A feature extractor extracts feature quantities from a digitized speech signal and outputs the feature quantities to a likelihood calculator. A distance determiner determines the distance between a user providing speech and a speech input unit. The likelihood calculator selects registered expressions for speech recognition from a recognition target table based on the determined distance, to be used in calculation of likelihoods at the likelihood calculator. The likelihood calculator calculates likelihoods for the selected registered expressions based on the feature quantities extracted by the feature extractor, and outputs one of the registered expressions having the maximum likelihood as a result of speech recognition.