US-12627766-B2 - Systems and methods for an artificial intelligence (AI) language model-enabled interactive narrative engine

Abstract

Embodiments of the disclosure can include systems and methods for an AI language model-enabled interactive narrative engine. In one embodiment, a system can comprise an interactive narrative engine with a voice mode manager module operable to receive a user input including at least one user intent; a service module operable to convert the user input for processing by a natural language processing (NLP) module, wherein based at least in part on the user input, the NLP module passes the resultant processed text to the user command preprocessing switch module, which determines at least one outcome for using the user input as an internal program command or as text in the fiction or non-fiction story or narrative; and an AI language model prompt manager operable to provide the at least one outcome to an AI language model, wherein based at least in part on the at least one outcome, the AI language model provides a response to the user's intent and facilitates narrating or telling the story or narrative.

Inventors

• THOMAS REALE

Assignees

• THOMAS REALE

Dates

Publication Date

20260512

Application Date

20230616

Claims (18)

1. 1 . A system for providing AI-generated narration content in a story with dynamically-generated content in response to an input from a user, the system comprising an interactive narrative engine comprising: a voice mode manager module operable to: receive a user input wherein the user identifies as a predefined character to play in the story; and receive a user input comprising at least one user intent to interact, speak, or move as the predefined character played by the user in the dynamically-generated content in the story; one or more modules operable to convert the user input for processing to discern the at least one user intent, wherein based at least in part on the user input and at least in part on output comprising the at least one user intent from a natural language processing (NLP) module, a user command preprocessing switch module determines at least one outcome for using the user input as an internal program command or as text for the user playing the predefined character in the dynamically-generated content in the story; and an AI language model prompt manager operable to provide the at least one outcome to an AI language model, wherein based at least in part on the at least one outcome, the AI language model provides a response to the at least one user intent and facilitates narrating or telling the dynamically-generated content with the user playing the predefined character in the story; and wherein the dynamically-generated content with the user playing the predefined character in the story is output to the user via an output device.
2. 2 . The system of claim 1 , wherein the user input comprises a voice input, a text input, a keyboard input, a mouse input, or a gesture input.
3. 3 . The system of claim 1 , wherein the at least one outcome comprises at least one of the following: issuing a command within the story that causes the user's character in the dynamically-generated content in the story to interact with an object in the story, or to interact with another character in the story; speaking as the user's character in the story with another character; or moving as the user's character within the story.
4. 4 . The system of claim 1 , wherein further based at least in part on the at least one outcome: the AI language model receives the user's responses to prompts in order to generate one or more character profiles for a new story or narrative; or the AI language model receives the user's responses to prompts in order to generate one or more character profiles, which are inserted into an existing story or narrative, which is modified so that one or more new characters interact with one or more characters already extant.
5. 5 . The system of claim 1 , further comprising: an AI image generator operable to receive the response to the user's intent from the AI language model; and based at least in part on the response, the AI image generator is operable to output one or more images.
6. 6 . The system of claim 1 , further comprising: an AI video generator operable to receive the response to the user's intent from the AI language model; and based at least in part on the response, the AI video generator is operable to output one or more video segments.
7. 7 . The system of claim 1 , wherein the AI language model comprises gpt-4, gpt-3.5-turbo, text-davinci-003, or another large language model.
8. 8 . A computer-implemented method for providing AI-generated content in a story with dynamically-generated content for a user to play a predefined character in the story, the method comprising: providing an interactive narrative engine operable to: receive a user input comprising at least one user intent to interact, speak, or move as a predefined character played by the user in the dynamically-generated content in the story; convert the user input for processing to discern the at least one user intent, wherein based at least in part on the user input and at least in part on the at least one user intent, determining at least one outcome for using the user input as an internal program command or as text in the dynamically-generated content for the user playing the predefined character in the story; provide the at least one outcome to an AI language model, wherein based at least in part on the at least one outcome, providing a response to the at least one user intent and facilitate narrating or telling the dynamically-generated content for the user playing the predefined character in the story; and outputting the dynamically-generated content for the user playing the predefined character in the story via an output device.
9. 9 . The computer-implemented method of claim 8 , wherein the user input comprises a voice input, a text input, a keyboard input, a mouse input, or a gesture input.
10. 10 . The computer-implemented method of claim 8 , wherein the at least one outcome comprises at least one of the following: issuing a command within the dynamically-generated content in the story that causes the user's character in the dynamically-generated content in the story to interact with an object in the dynamically-generated content in the story, or to interact with another character in the dynamically-generated content in the story; speaking as the user's character in the dynamically-generated content in the story with another character; or moving as the user's character within the dynamically-generated content in the story.
11. 11 . The computer-implemented method of claim 8 , further comprising: receive the response to the user's intent from the AI language model; and based at least in part on the response, output one or more images.
12. 12 . The computer-implemented method of claim 8 , further comprising: receive the response to the user's intent from the AI language model; and based at least in part on the response, output one or more video segments.
13. 13 . The computer-implemented method of claim 8 , wherein the AI language model comprises gpt-4, gpt-3.5-turbo, text-davinci-003, or another large language model.
14. 14 . A non-transitory computer-readable medium comprising computer-executable instructions for: receiving a user input comprising at least one user intent to interact, speak, or move as a predefined character played by the user in dynamically-generated content in a story or narrative; converting the user input for processing to discern the at least one user intent, wherein based at least in part on the user input and at least in part on the at least one user intent, determining at least one outcome for using the user input as an internal program command or as text in the dynamically-generated content for the user playing the predefined character in the story or narrative; providing the at least one outcome to an AI language model comprising gpt-4 or another large language model, wherein based at least in part on the at least one outcome, providing a response to the at least one user intent and facilitating narrating or telling the dynamically-generated content for the user playing the predefined character in the story or narrative; and outputting the dynamically-generated content for the user playing the predefined character in the story or narrative via an output device, and wherein the at least one outcome comprises at least one of the following: issuing a command within the dynamically-generated content in the story or narrative that causes the user's character in the dynamically-generated content in the story or narrative to interact with an object in the dynamically-generated content in the story or narrative, or to interact with another character in the dynamically-generated content in the story or narrative; speaking as the user's character in the dynamically-generated content in the story or narrative with another character; or moving as the user's character within the dynamically-generated content in the story or narrative.
15. 15 . The computer-readable medium of claim 14 , wherein the user input comprises a voice input, a text input, a keyboard input, a mouse input, or a gesture input.
16. 16 . The computer-readable medium of claim 14 , further comprising computer-executable instructions for: receiving the response to the user's intent from the AI language model; and based at least in part on the response, outputting one or more images.
17. 17 . The computer-readable medium of claim 14 , further comprising computer-executable instructions for: receiving the response to the user's intent from the AI language model; and based at least in part on the response, outputting one or more video segments.
18. 18 . The computer-readable medium of claim 14 , wherein further based at least in part on the at least one outcome: the AI language model receives the user's responses to prompts in order to generate one or more character profiles for a new story or narrative; or the AI language model receives the user's responses to prompts in order to generate one or more character profiles, which are inserted into an existing story or narrative, which is modified so that one or more new characters interact with one or more characters already extant.

Description

RELATED APPLICATIONS The present application claims priority to U.S. Ser. No. 63/453,095, filed Mar. 18, 2023, and U.S. Ser. No. 63/455,038, filed Mar. 28, 2023, the contents of which are both incorporated by reference. TECHNICAL FIELD The present disclosure relates to artificial intelligence (AI), and more particularly relates to, systems and methods for an AI language model-enabled interactive narrative engine. BACKGROUND Conventional application programs that purport to provide interactive narration of a story or a book have limited support for language processing, and no support for voice participation. As a result, the interactive narration is forced and does not allow for truly “free play.” Further, certain application programs can utilize conventional AI-language processing for storytelling, however, these application programs are typically focused on playing through static, pre-existing human-authored content that does not allow user-influenced dynamic content generation within a human-authored story. In either instance, the user experience with conventional interactive narration application programs and with certain application programs utilizing conventional AI-language processing for storytelling can lead to a greatly diminished user experience. This is because conventional interactive narration programs are deterministic in nature, where playing through a story leads to the same exact outcome, or to a limited number of exactly identical outcomes with each “reading.” Some users may lose their place in the story or book, other users may lose interest in the story or book, and yet other users may lose confidence in using conventional interactive narration application programs and certain application programs utilizing conventional AI-language processing for storytelling. SUMMARY The present disclosure addresses a need for systems and methods for an AI language model-enabled interactive narrative engine. Applications for the embodiments within the disclosure include, but are not limited to, entertainment through literature, access to AI-enabled interactive narratives for the blind and visually impaired, and supporting literacy development within education systems including early literacy, reading remediation and intervention, and second-language acquisition including English as a second language. Multimodal media support for the embodiments include “Book mode,” “Theater mode,” and “Film mode,” where user input is interpreted by the interactive narrative engine, and a series of AI language model responses will generate, or supplement, a fiction or non-fiction story or narrative in the target media per the user's intent. Embodiments of the disclosure can include systems and methods for an AI language model-enabled interactive narrative engine. In one embodiment, a system can include an interactive narrative engine with a voice mode manager module operable to receive a user input including at least one user intent; one or more modules operable to convert the user input for processing to discern the at least one user intent, wherein based at least in part on the user input and at least in part on output comprising the at least one user intent from a natural language processing (NLP) module, a user command preprocessing switch module determines at least one outcome for using the user input as an internal program command or as text in the fiction or non-fiction story or narrative; and an AI language model prompt manager operable to provide the at least one outcome to an AI language model, wherein based at least in part on the at least one outcome, the AI language model provides a response to the at least one user intent and facilitates narrating or telling the story or narrative. In at least one aspect of one embodiment, the user input can include a voice input, a text input, a keyboard input, a mouse input, or a gesture input. In at least one aspect of one embodiment, the at least one outcome can include at least one of the following: issuing a command within the fiction or non-fiction story or narrative that causes the user's character in the story or narrative to interact with an object in the story or narrative, or to interact with another character in the story or narrative; speaking as the user's character in the story or narrative with another character; or moving as the user's character within the story or narrative. In at least one aspect of one embodiment, wherein further based at least in part on the at least one outcome: the AI language model receives the user's responses to prompts in order to generate one or more character profiles for a new story or narrative; or the AI language model receives the user's responses to prompts in order to generate one or more character profiles, which are inserted into an existing story or narrative, which is modified so that one or more new characters interact with one or more characters already extant. In at least one aspect of one embodiment, the system fu