Methods for examining game context for determining a user's voice commands
A method for executing a session of a video game is provided, including the following operations: recording speech of a player engaged in gameplay of the session of the video game; analyzing a game state generated by the execution of the session of the video game, wherein analyzing the game state identifies a context of the gameplay; analyzing the recorded speech using the identified context of the gameplay and a speech recognition model, to identify textual content of the recorded speech; applying the identified textual content as a gameplay input for the session of the video game.
1. A method for executing a session of a video game, comprising:
recording speech of a player engaged in gameplay of the session of the video game;
analyzing the recorded speech using a speech recognition model, which identifies a plurality of candidate words as possible interpretations of the speech of the player; during the session of the video game, presenting the plurality of candidate words to the player;
receiving selection input from the player identifying one of the candidate words as a correct interpretation of the speech of the player; and
applying the selected one of the candidate words by the video game as a gameplay input for the video game.
2. The method of claim 1 , wherein presenting the plurality of candidate words is responsive to a level of confidence of recognition by the speech recognition model falling below a predefined threshold.
3. The method of claim 1 , wherein presenting the candidate words to the player includes presenting the candidate words in video generated from the session of the video game.
4. The method of claim 1 , wherein presenting the candidate words pauses the gameplay until the selection input has been received.
5. The method of claim 1 , wherein applying the selected one of the candidate words includes triggering a command for the gameplay of the video game.
6. The method of claim 1 , wherein the selected one of the candidate words is used as feedback to refine the speech recognition model.
7. The method of claim 1 , wherein receiving selection input is through an input device of a controller.