In-Vehicle Speech Recognition Correction via Steering Wheel Touchpad

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional speech recognition systems face challenges in accurately correcting misrecognized speech content, especially in noisy environments, where similar correction instructions can lead to repeated misrecognition, causing user stress.

Innovation Solution

A speech recognition apparatus with touch pads attached to a steering wheel, allowing manual correction of misrecognized speech content through tapping or swiping operations, which generates a correction character string to replace the misrecognized characters, improving accuracy and user understanding.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If speech recognition is performed in noisy environments, then the system can operate in running states, but recognition accuracy deteriorates

Engineering Contradiction:

Improveoperation in running stateVSAvoidspeech recognition accuracy

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent introduces an operation portion as an intermediary input method between the user and the speech recognition system. When speech recognition accuracy deteriorates in noisy environments, the user can switch to manual input through the operation portion (touchpad or buttons) to provide correction instructions or alternative input, thereby compensating for the unreliable speech recognition without requiring the system to perfectly interpret noisy speech

Inventive Principle:

Principle #24Intermediary (Mediator)

2Ease of operation

If correction instructions are provided through voice, then correction can be performed, but misrecognition occurs when making voices resemble each other

Engineering Contradiction:

Improvecorrection capabilityVSAvoidcorrection instruction recognition accuracy

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The operation portion serves as a reliable intermediary input method that does not suffer from the limitations of voice-based correction. When voice correction leads to misrecognition due to similar-sounding instructions, the user can use the operation portion (touchpad or buttons) to input correction instructions manually, providing a precise alternative that is not affected by acoustic similarities

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system allows the user to self-correct recognition errors by providing direct manual input capability through the operation portion. Instead of relying on the system to interpret potentially ambiguous correction voices, the user can directly input corrections through tactile interaction, giving them control over the correction process and eliminating misrecognition issues

Inventive Principle:

Principle #25Self-service

3Adaptability or versatility

If multiple correction patterns are displayed, then correction options are provided, but user stress increases when repeated misrecognition occurs

Engineering Contradiction:

Improvecorrection optionsVSAvoiduser stress level

Core Design Contradiction:

Adaptability or versatilityVSEase of operation

Solution Approach 1:

The operation portion provides a reliable direct input channel that bypasses the speech recognition system's limitations. When repeated misrecognition occurs despite displaying multiple correction patterns, the user can use the operation portion to input corrections manually, gaining certainty that their input will be accurately registered without the frustration of voice misrecognition

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system empowers the user to take direct control of the correction process through the operation portion. Instead of passively selecting from displayed correction patterns that may be misrecognized, the user can actively input their intended correction through tactile interaction, reducing stress by having reliable control over the correction outcome

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS9153234B2Speech recognition apparatus, method of recognizing speech, and computer readable medium for the same

Publication Date: 2015.10.06 DENSO CORP

AI summary

A speech recognition apparatus includes: a recognition device that recognizes a speech of a user and generates a speech character string; a display device that displays the speech character string; a reception device that receives an input of a correction character string, which is used for correction of the speech character string, through an operation portion; and a correction device that corrects the speech character string with using the correction character string.