Contextual Language Translation Using Nested Text Strings

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Automated translation systems face difficulties in translating words without sufficient contextual information, particularly in languages where nouns have gender, leading to inaccuracies in adjective agreement with modified nouns.

Innovation Solution

The system receives a first text string, generates a translation, and then combines it with subsequent text to utilize contextual information such as gender and verb tenses for more accurate translations, removing previously translated portions to provide only new translations to the user.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the automated translation system translates each text string independently without context, then the translation process is simple and fast, but the translation accuracy deteriorates due to insufficient contextual information

Engineering Contradiction:
Improvetranslation accuracyVSAvoidtranslation process complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system performs preliminary translation of the first text string and stores it along with its translation. This preliminary action provides contextual information that will be used when translating subsequent text strings, allowing the system to maintain accuracy without re-translating entire contexts.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system nests the first text string and its translation within the combined text string context. When translating the second text string, the system incorporates the previously translated content as nested contextual information, enabling accurate gender and tense agreement without processing the entire text from scratch.

Inventive Principle:
Principle #7Nested doll (Nesting)

2Ease of operation

If the automated translation system provides all previously translated text to the user, then the user has complete context, but the user interface becomes cluttered and less efficient

Engineering Contradiction:
Improveuser interface efficiencyVSAvoidcontextual information loss
Core Design Contradiction:
Ease of operationVSLoss of information

Solution Approach 1:

The system extracts only the translation of the second text string from the combined translation output and presents it to the user. The previously translated first text string is excluded from the user interface output, eliminating clutter while maintaining translation accuracy through contextual processing.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system acts as an intermediary between the full combined translation and the user interface. It processes the complete contextual translation but selectively outputs only the new translation portion, serving as a mediator that filters information for optimal user presentation.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Measurement precision

If the automated translation system processes each text string separately, then processing speed is fast, but translation accuracy deteriorates due to lack of contextual information for gender and verb tense agreement

Engineering Contradiction:
Improveadjective agreement accuracyVSAvoidtranslation processing efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The system performs preliminary translation of the first text string and stores contextual information including gender and verb tense. This preliminary processing enables accurate adjective agreement in subsequent translations without requiring re-processing of the entire context, maintaining both accuracy and efficiency.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system merges the first text string with the second text string to form a combined text string for translation. This merging allows the translation engine to process contextual information together while still maintaining efficient incremental output, resolving the contradiction between comprehensive processing and processing speed.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS10229114B2Contextual language translation
Publication Date: 2019.03.12 GOOGLE LLC
  • US10229114B2 patent drawing
  • US10229114B2 patent drawing
  • US10229114B2 patent drawing

AI summary

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for contextual language translation. In one aspect, a method includes the actions of receiving a first text string. The actions further include generating a first translation of the first text string. The actions further include providing, for output, the first translation of the first text string. The actions further include receiving a second text string. The actions further include generating a combined text string by combining the first text string and the second text string. The actions further include generating a second translation of the combined text string. The actions further include providing, for output, a portion of the second translation that includes a translation of the second text string without providing, for output, a portion of the second translation that includes a translation of the first text string.