Electronic Apparatus Sentence Separation for Translation Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing translation systems face challenges in achieving accurate translations when processing multiple words from images without proper sentence separation, leading to decreased translation accuracy.
Innovation Solution
An electronic apparatus and method that separates multiple words into sentence units using language models and sentence separation models to improve translation accuracy by inputting word combinations to a translation model.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If all words are input into translation model without sentence separation, then processing is simple, but translation accuracy decreases
Solution Approach 1:
The patent segments the input text into sentence units before translation. The sentence separation model divides the continuous text stream into discrete sentences, which are then processed individually by the translation model. This segmentation enables the translation system to handle each sentence independently, improving translation accuracy while maintaining manageable processing complexity through automated sentence boundary detection.
2Manufacturing precision
If sentence separation is performed, then translation accuracy improves, but processing time increases
Solution Approach 1:
The patent performs sentence separation as a preliminary action before translation. The sentence separation model pre-processes the input text to identify and separate sentences, creating ready-to-process sentence units for the translation model. This preliminary segmentation prepares the data in an optimized format that reduces subsequent translation processing time and improves accuracy.
Solution Approach 2:
The sentence separation model acts as an intermediary between the input text and the translation model. It mediates the processing by transforming the raw text into structured sentence units, which serves as an optimal interface for the translation model. This intermediary layer enables efficient data transformation without requiring the translation model to handle raw text directly, thereby improving both accuracy and processing efficiency.
Data Source
AI summary
An electronic apparatus and a control method thereof are provided. The control method includes acquiring a plurality of words, acquire a plurality of word combinations by combining the plurality of words, and acquiring first scores indicating the sentence completeness of each of the plurality of word combinations by inputting each of the plurality of word combinations to a first a language model for identifying a sentence completeness, based on all the first scores with respect to each of the plurality of word combinations being identified to be less than a predetermined value, acquire second scores for each of the plurality of word combinations by inputting each of the plurality of word combinations to a sentence separation model for identifying an accuracy of a word combination, identifying one word combination among the plurality of word combinations based on the second scores, and providing the identified one word combination to a translation model which performs translation.