Back to results
US 2026/0145064 A1Published application
Smart microphone to improve detection of speaker
AssigneeSony Interactive Entertainment Inc.
InventorsLakshmish Kaushik, Xiaoyong Ye, Mario Sarria, Koichi Obana, Sergey Bashkirov, Michael Prosinski
Life of this patent
Estimating the term…
Abstract
Non-player characters (NPC) in a computer game may be controlled using context in the form of player statements from the game engine to an LLM to drive NPC dialog and control NPC actions. Imperative statements are executed by game engine, and the result of this execution plus the original player request are converted to plain text description and injected back into the NPC dialog to ensure the NPC remains within defined bounds.
Claims20
An apparatus comprising:
- at least one processor system configured to:
- receive data representing speech from a player of a computer game;
- and
- generate dialog of at least one non-player character (NPC) in the computer game based at least in part on the data representing speech from the player.
1 more element
An apparatus comprising:
- computer memory that is not a transitory signal and that comprises instructions executable by at least one processor system to:
- semantically analyze statements from a player of a computer game using one or more language models to generate dialog of a non-player character (NPC) of the computer game;
- execute imperative statements from the player using a game engine associated with the computer game;
- convert into text a result of executing imperative statements from the player to generate NPC dialog to ensure the NPC dialog remains synchronized with behavior of the NPC.
1 more element
A method, comprising:
- controlling action of a non-player character (NPC) in a computer game;
- controlling dialog in the NPC;
- wherein the controlling steps are executed using artificial intelligence (AI) based on NPC speech generation based on context, semantic analysis of player statements to convert to machine readable format, which a game engine executes and also sends text of to a language model to give the model context.
Dependencies are inferred from the claim text.