Machine Translation Network for Linguistic Variations

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional machine translation systems fail to properly account for linguistic variations and dialects in specific corpora, leading to a loss of true meaning in translated text.

Innovation Solution

A machine learning network is trained using human-augmented translations from specialized communities, incorporating linguistic variations and dialects to generate more accurate translations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional machine translation systems are used, then standard translation is provided, but linguistic variations and dialects are not properly accounted for, leading to loss of true meaning

Engineering Contradiction:

Improvetranslation accuracyVSAvoidlinguistic variations and dialects

Core Design Contradiction:

ReliabilityVSLoss of information

Solution Approach 1:

The patent segments the translation task by creating separate processing paths: conventional machine translation for standard text and a specialized machine learning network for detecting and translating linguistic variations and dialects. This segmentation allows each component to handle specific aspects of translation, improving overall accuracy while preserving linguistic nuances.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary component - a machine learning network trained on human-augmented translations - that acts as a mediator between the input text and the final translation. This intermediary detects linguistic variations and dialects that conventional systems miss, bridging the gap between standard translation and accurate representation of true meaning.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Adaptability or versatility

If conventional machine translation systems are used, then translation processing is simple, but translations fail to account for linguistic variations present in specific corpora

Engineering Contradiction:

Improvehandling of linguistic variationsVSAvoidtranslation system complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The translation system is segmented into conventional machine translation components and a specialized machine learning network. This segmentation allows the system to maintain simplicity for standard translations while adding adaptability for linguistic variations only where needed, rather than redesigning the entire system.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The machine learning network is pre-trained on human-augmented translations that incorporate linguistic variations and dialects. This preliminary action equips the system with the ability to handle diverse linguistic forms before actual translation tasks, enabling adaptability without increasing operational complexity during translation processing.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11640233B2Foreign language machine translation of documents in a variety of formats

Publication Date: 2023.05.02 VANNEVAR LABS INC

AI summary

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for extracting text from an input document to generate one or more inference. Each inference box may be input into a machine learning network trained on training labels. Each training label provides a human-augmented version of output from a separate machine translation engine. A first translation may be generated by machine learning network. The first translation may be displayed in a user interface with respect to display of an original version of the input document and a translated version of a portion of the input document.