Voice Recognition Numeral Digit Limit Handling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Voice recognition systems often misinterpret numerals when spoken with pauses or enunciations, leading to errors in recognizing numerical values with more digits than the system's preset maximum, resulting in incorrect translations and outputs.
Innovation Solution
A voice recognition apparatus that acquires a speech sentence, checks the number of digits in a numeral, and if it exceeds the maximum allowed, replaces it with a value composed of multiple numerals that sum to the original, ensuring the output has digits equal to or less than the maximum, thereby correcting errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the voice recognition system uses a preset maximum digit limit, then it can prevent recognition errors for standard cases, but it fails to correctly recognize numerical values exceeding this limit
Solution Approach 1:
The patent segments a long numeral exceeding the maximum digit limit into multiple smaller numerals. For example, a 6-digit number is divided into two 3-digit numbers. These segmented numerals are then processed separately and combined through addition to reconstruct the original value, enabling accurate recognition of numbers beyond the system's preset digit limit while maintaining compatibility with standard maximum-length constraints.
2Manufacturing precision
If the system replaces long numerals with multiple shorter numerals, then it can output within digit limits, but it requires additional processing steps to divide and sum the numerals
Solution Approach 1:
The patent performs preliminary segmentation of the numeral into smaller units before processing. By pre-dividing the long numeral into manageable segments that fit within the maximum digit limit, the system prepares the data in advance for accurate processing. This preliminary action simplifies subsequent operations, as the segmented numerals can be directly added together without requiring complex real-time processing during output generation.
Data Source
AI summary
A voice recognition apparatus includes: a communication circuit acquiring a speech sentence that is a result of voice recognition of a speech; a storage storing digit number information indicative of the maximum number of digits; and a control circuit. When the number of digits of a first numerical value indicated by a first numeral included in the speech sentence is larger than the maximum number of digits, the control circuit replaces the first numeral in the speech sentence with a second numeral indicative of a second numerical value having the number of digits equal to or less than the maximum number of digits. The control circuit divides the first numeral into a plurality of numerals and adds numerical values respectively indicated by the plurality of numerals to calculate the second numerical value.