Hybrid Voice-Handwriting Recognition for Input Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current handwriting recognition systems face challenges in accurately converting handwriting input into machine text due to the complexity of individual handwriting styles and extensive languages, leading to errors and inefficiencies in information entry.
Innovation Solution
The integration of voice input recognition with handwriting recognition, where both inputs are processed to generate machine input word lists, allowing for the determination of a highest probability word, thereby enhancing the accuracy and efficiency of character recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If handwriting recognition is used to convert handwriting input into machine text, then users can write more naturally without a keyboard, but recognition accuracy deteriorates due to complexity of individual handwriting styles and extensive languages
Solution Approach 1:
The patent combines handwriting recognition with voice recognition to form a hybrid input system. The speech recognition component provides contextual information and probability scores that help disambiguate handwritten input, thereby improving overall recognition accuracy while maintaining the natural handwriting input method.
Solution Approach 2:
The system uses voice input as feedback to enhance handwriting recognition. By comparing speech recognition results with handwriting recognition results, the system can confirm or correct recognized text, providing a feedback mechanism that improves accuracy without requiring users to switch input methods.
2Measurement precision
If handwriting recognition processes consecutive characters to generate candidate words, then more context is available for recognition, but processing time increases and productivity decreases
Solution Approach 1:
The system performs preliminary speech recognition on spoken input while the user is still writing or immediately after. This preliminary processing provides candidate words and contextual information in advance, reducing the computational burden on handwriting recognition and enabling faster overall processing without sacrificing accuracy.
Solution Approach 2:
By merging speech recognition results with handwriting recognition results, the system can quickly resolve ambiguities without extensive processing of consecutive handwriting characters. The speech component provides immediate contextual clues that reduce the need for complex sequential handwriting analysis.
Data Source
AI summary
One embodiment provides a method, including: receiving, at an input and display device, handwriting input; receiving, using a processor, voice input; generating, using a processor, at least one first word based on the handwriting input; generating, using a processor, at least one second word based on the voice input; and determining, using a processor, a highest probability word based on the at least one first word and the at least one second word. Other aspects are described and claimed.