Back to results
US 7,734,562 B1Granted
Voice to text conversion with keyword parse and match to semantic and transactional concepts stored in a brain pool state machine using word distance to generate character model interaction in a plurality of dramatic modes
AssigneeBrainpool, Inc.
InventorRobert Hairman
Life of this patent
Estimating the term…
Abstract
A system and method is disclosed that supervises the behavior of interactive characters using dynamic, fuzzy, social rules. These characters, called Protags, resemble human beings and are modeled after universal characters and story motifs. Protags respond to input from speech recognizers, text based natural language recognition systems and manipulation of objects in the virtual world in which these characters reside.
Claims11
A computer-implemented method comprising:
- at a computer, receiving conversational voice input from a user;
- with a processor of the computer, converting the conversational voice input to a text string;
- parsing the text string representation of the received input;
- providing in the computer a repository of key words;
- matching patterns in the text string to key words to semantic and transactional concepts using word distance, wherein the concepts reside in a brain pool state machine, implemented using computer code executing on the computer, the computer code residing on a mass storage device of the computer;
- and
- providing an interaction output, via a computer monitor and speaker coupled to the computer, to the user based on the brain pool state machine.
4 more elements
A computer-implemented method of creating an interactive character comprising:
- at a computer, receiving a conversational voice input from a user;
- with a processor of the computer, converting the conversational voice input to a text string;
- parsing the text string representation of the received input;
- matching patterns in the text string to key words to semantic concepts using word distance according to a character model for the interactive character;
- implementing in the character model a plurality of dramatic modes, wherein each dramatic mode has a parameter specifying when that dramatic mode is in effect for a certain time period in a timeline of a story;
- when the story is in a first time period, selecting a first dramatic mode where the conversational voice input from a user has a first percentage chance of invoking a mode change of the character;
- when the story is in a second time period after the first time period, selecting a second dramatic mode where the conversational voice input from a user has a second percentage chance of invoking the mode change of the character, wherein the second percentage chance is greater than the first percentage chance;
- and
- outputting to a display screen and speaker of the computer a representation of the character being in the changed mode in the first time period when the first percentage chance is exceeded and in the second time period when the second percentage chance is exceeded.
6 more elements
Dependencies are inferred from the claim text.