Voice Input Error Correction via Keypad Reference

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current voice input systems in electronic devices often suffer from inaccuracies, leading to errors in text recognition, which can result in incorrect conversions and reduced user experience.

Innovation Solution

A method and apparatus that involve receiving a first word input via a keypad, recognizing voice input, converting it into text, and correcting errors by replacing incorrectly recognized words with the original keypad input, thereby improving text input accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If voice recognition is used for text input, then input speed and convenience are improved, but recognition accuracy deteriorates

Engineering Contradiction:

Improveinput speedVSAvoidrecognition accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The system compares the voice-recognized text with the originally input text and automatically corrects errors by replacing mismatched words. This feedback mechanism continuously improves recognition accuracy by learning from correction patterns while maintaining fast voice input speed.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system stores the originally input text before voice recognition occurs. This preliminary action creates a reference baseline that enables automatic error correction by comparing recognized text against the known correct input, thereby improving accuracy without slowing down the input process.

Inventive Principle:

Principle #10Preliminary action

2Ease of operation

If voice recognition is used for text input, then ease of operation is improved, but error rate increases

Engineering Contradiction:

Improveinput convenienceVSAvoiderror rate

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The system implements automatic error correction by comparing voice-recognized text with the original input and replacing erroneous words. This feedback loop maintains high ease of operation through voice input while significantly reducing the error rate through automated verification and correction.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system prepares correction data by storing the originally input text before voice recognition occurs. This beforehand cushioning creates a safety net that cushions against recognition errors, allowing the system to automatically recover from mistakes without requiring manual intervention, thus maintaining both convenience and reliability.

Inventive Principle:

Principle #11Beforehand cushioning (Prior cushioning)

Data Source

PatentUS9583103B2Method of controlling a text input and electronic device thereof

Publication Date: 2017.02.28 SAMSUNG ELECTRONICS CO LTD

AI summary

According to various embodiments, a method for an electronic device includes receiving an input of a first word from a keypad, recognizing a voice input and converting the voice input into a text including a second word, in response to determining that the second word is erroneously recognized based on the first word, correcting the text by replacing the second word with the first word, and entering the corrected text. An electronic device for recognizing a voice includes a keypad configured to receive an input of a first word, a sensor configured to recognize a voice input, a controller configured to convert the voice input into a text including a second word, in response to determining that the second word is erroneously recognized based on the first word, correct the text by replacing the second word with the first word, and enter the corrected text.