Automated Game Script Generation via NLP Control Data Inference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The development of game scripts for narrativity-based games requires significant manual intervention, leading to high costs and labor, with a need for automation to reduce manual input without compromising quality.
Innovation Solution
A system that includes a data pre-processing module to convert control data into natural language explanation text and a learning module using a pre-trained natural language model to generate trained models for automating game script creation, allowing for the automatic generation of game scripts by inferring control explanation text from input explanation text.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual entry of entire game scripts is performed, then quality of game script can be maintained, but development cost and labor increase significantly
Solution Approach 1:
The patent segments the game script into two parts: explanation text (natural language) and control data (structured commands). The system automatically generates control data from explanation text, while manual input is required only for explanation text. This segmentation allows automation of the labor-intensive control data generation while preserving quality through manual review of explanation text.
Solution Approach 2:
The patent replaces the mechanical manual entry process with an AI-based natural language processing system. The learning module uses trained models to automatically convert explanation text into control data, substituting manual typing and copying operations with automated machine learning inference.
2Productivity
If automatic generation of game scripts is implemented, then manual labor is reduced, but quality may be compromised
Solution Approach 1:
The system incorporates feedback mechanisms where the learning module continuously improves by reviewing automatically generated control data against actual game execution results. This feedback loop allows the system to learn from errors and improve quality over time, reducing the need for complete manual rewriting.
Solution Approach 2:
The patent introduces explanation text as an intermediary representation between the developer's intent and the machine-executable control data. This intermediary layer allows for easier automated conversion while maintaining interpretability, as the explanation text serves as a human-readable bridge that can be reviewed and corrected if needed.
3Extent of automation
If control data is converted to natural language explanation text, then automation becomes possible, but data processing complexity increases
Solution Approach 1:
The patent performs preliminary conversion of control data into explanation text during the data preparation phase. This preliminary action creates training data in a format that is easier for AI models to process, reducing the complexity of subsequent automated generation while maintaining the ability to accurately represent control intentions.
Data Source
AI summary
One or more embodiments of the invention provides a system for supporting the creation of a game script including natural language data representing explanation text in a game and also including control data for controlling the game, the natural language data and the control data being associated in accordance with the content of the game, the system including: a data pre-processing module that converts control data included in created game scripts created in advance into control explanation text in the form of natural language data and creates processed script text including the explanation text and control explanation text corresponding to the explanation text; and a learning module that generates a trained model by causing a pre-trained natural language model to learn the processed script text, the pre-trained natural language model having learned in advance grammatical structures and text-to-text relationships concerning natural language text.