Voice Recognition Apparatus Activation Phrase Exclusion

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Users experience voice instruction recognition errors when repeating activation phrases in voice interaction systems, leading to incorrect recognition of voice commands.

Innovation Solution

A voice recognition apparatus and method that includes an input unit, a control unit, and a recognition unit to accept and process voice inputs, where the recognition unit excludes the activation phrase from the voice input after the system has been activated, ensuring accurate recognition of voice commands by comparing voice data with stored feature amounts and adjusting thresholds based on the length of the activation phrase.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the system continuously monitors for activation phrases to enable voice instruction recognition, then the system remains responsive to user commands, but the system may incorrectly recognize repeated activation phrases as new voice instructions

Engineering Contradiction:
Improvevoice instruction recognition accuracyVSAvoiduser anxiety about voice input
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The system performs preliminary action by detecting the activation phrase before processing the voice instruction. When an activation phrase is detected, the system sets a flag to indicate that the subsequent voice input should be processed as an instruction. This preliminary detection prevents the activation phrase itself from being misinterpreted as a command, resolving the contradiction between maintaining responsiveness and preventing recognition errors.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The voice processing is segmented into distinct phases: activation phrase detection phase and instruction processing phase. The system divides the continuous voice monitoring into discrete segments where the first segment identifies the activation trigger and the second segment processes the actual command. This segmentation allows the system to distinguish between activation phrases and instructions, improving recognition accuracy while maintaining ease of operation.

Inventive Principle:
Principle #1Segmentation

2Speed

If the system accepts voice input immediately after activation phrase detection, then the response time is minimized, but the system may process incomplete or erroneous voice instructions

Engineering Contradiction:
Improvevoice instruction processing speedVSAvoidvoice instruction recognition accuracy
Core Design Contradiction:
SpeedVSReliability

Solution Approach 1:

The system performs preliminary voice quality assessment and completeness checking before final processing. After detecting the activation phrase and receiving the voice instruction, the system conducts a preliminary analysis of the voice signal characteristics and instruction structure. This preliminary action allows rapid filtering of obviously erroneous inputs while preserving fast processing of valid commands, thus maintaining both speed and reliability.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS11164578B2Voice recognition apparatus, voice recognition method, and non-transitory computer-readable storage medium storing program
Publication Date: 2021.11.02 HONDA MOTOR CO LTD
  • US11164578B2 patent drawing
  • US11164578B2 patent drawing
  • US11164578B2 patent drawing

AI summary

A voice recognition apparatus capable of executing a function by a voice instruction, comprises an input unit configured to input a voice, and a control unit configured to enable the voice recognition apparatus to accept a voice instruction for executing a function, if a voice input by the input unit is an activation phrase. The voice recognition apparatus comprises a recognition unit configured to, if a voice input by the input unit after the control unit has enabled the voice recognition apparatus to accept the voice instruction contains the activation phrase, recognize the voice instruction by excluding the activation phrase from the voice.