Method and apparatus for generating plot in miniature theater intelligent machine, device, medium, and product

A apparatus for generating a plot in a miniature theater intelligent machine, comprising:

• the miniature theater intelligent machine, configured to execute a method, wherein the miniature theater intelligent machine comprises a sensor and a loudspeaker, and the sensor is configured to sense an operation instruction of a user, and the loudspeaker is configured to play a sound effect in the plot;
• a far infrared induction flashlight, provided with a card slot, wherein
• the card slot is used for insertion of a character near field communication (NFC) card, and the character NFC card is used for storing character information, wherein the character information comprises personage dialogue, a scene corresponding to the personage dialogue, and personage information;
• and
• the far infrared induction flashlight is configured to read the character NFC card to trigger a corresponding plot, and irradiate a personage on a screen of the miniature theater intelligent machine by a far infrared ray to trigger a plot text corresponding to the personage;
• a projector, configured to project a plot scene corresponding to a plot text at the next moment generated by the miniature theater intelligent machine onto the screen;
• and
• a power module, configured to supply power to the miniature theater intelligent machine, the far infrared induction flashlight, and the projector;
• wherein the method comprises:
• performing word segmentation on a plot text at a current moment to generate a plurality of words, wherein the plot text comprises an educational story or game plotline, a plot category and personage dialogue corresponding to the plot category;
• transforming the plurality of words by Bidirectional Encoder Representations from Transformers to generate word vectors corresponding to respective words;
• setting the word vectors corresponding to the respective words as feature vectors corresponding to the plot text at the current moment;
• and
• inputting the feature vectors corresponding to the plot text at the current moment and contextual information at the current moment into a trained deep learning model to determine the plot text at the next moment, wherein the contextual information at the current moment comprises a historical choice of the user at the current moment, an emotional state of a character at the current moment, and a real-time operation instruction of the user at the current moment, and the real-time operation instruction comprises a voice instruction and a touch instruction.

A computer device, comprising:

• a memory,
• a processor, and
• a computer program, stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements a method;
• wherein the method comprises:
• performing word segmentation on a plot text at a current moment to generate a plurality of words, wherein the plot text comprises an educational story or game plotline, a plot category and personage dialogue corresponding to the plot category;
• transforming the plurality of words by Bidirectional Encoder Representations from Transformers to generate word vectors corresponding to respective words;
• setting the word vectors corresponding to the respective words as feature vectors corresponding to the plot text at the current moment;
• and
• inputting the feature vectors corresponding to the plot text at the current moment and contextual information at the current moment into a trained deep learning model to determine a plot text at a next moment, wherein the contextual information at the current moment comprises a historical choice of a user at the current moment, an emotional state of a character at the current moment, and a real-time operation instruction of the user at the current moment, and the real-time operation instruction comprises a voice instruction and a touch instruction.