| CPC G10L 25/63 (2013.01) [G06F 40/30 (2020.01)] | 15 Claims |
|
1. A method for assisting a neurological disorders patient through a therapy session using a voice interaction with a narrative audio, wherein the neurological disorders patient is a person with neurological disorders (a patient), the method comprising:
extracting, by the hardware processor, emotional content from the set of multimedia in a playlist, wherein the playlist of content is created by an upstream therapy application;
calculating, by the hardware processor, an emotional score for each emotional content in the created playlist using a deep machine learning model, wherein the emotional content is related to at-least one of: happiness, sadness, surprise, fear, anger, disgust, and contempt;
generating, by the hardware processor, an emotional arc for each of the emotional content in the created playlist based on the calculated emotional score, a previous response from the patient to a plurality of genres and personal information of the patient, wherein the response to the plurality of genres comprises features that comprise various emotions and communications initiated by at-least one of: the patient and a caretaker of the patient;
generating, by the hardware processor, a narrative text from the set of multimedia in the created playlist and generating a narrative text story based on the generated narrative text and the personal information of the patient using an artificial intelligence (AI) story model;
converting, by the hardware processor, the generated narrative text story to an audio file;
generating, by the hardware processor, a narrative audio by combining the audio file from the text to speech conversion subsystem with the emotional arc, wherein the narrative audio is modulated with corresponding pitch and tone based on a value of the generated emotional arc; and
outputting, by the hardware processor, the generated narrative audio to the patient in a voice of a robot.
|
|
8. A system for assisting a neurological disorders patient through a therapy session using a voice interaction with a narrative audio, wherein the neurological disorders patient is a person with neurological disorders (a patient), the system comprising:
a hardware processor; and
a memory coupled to the hardware processor, wherein the memory comprises a set of program instructions in the form of a plurality of subsystems, configured to be executed by the hardware processor, wherein the plurality of subsystems comprises:
a content extraction subsystem that is configured to extract emotional content from the set of multimedia in a playlist, wherein the playlist of content is created by an upstream therapy application;
an emotional score calculation subsystem that is configured to calculate an emotional score for each emotional content in the created playlist using a deep machine learning model, wherein the emotional content is related to at-least one of: happiness, sadness, surprise, fear, anger, disgust, and contempt;
an emotional arc generation subsystem that is configured to generate an emotional arc for each of the emotional content in the created playlist based on the calculated emotional score, a previous response from the patient to a plurality of genres and personal information of the patient, wherein the response to the plurality of genres comprises features that comprise various emotions and communications initiated by at-least one of: the patient or and caretaker of the patient;
a narrative story generation subsystem that is configured to generate a narrative text from the set of multimedia in the created playlist and to generate a narrative text story based on the generated narrative text and the personal information of the patient using an artificial intelligence (AI) story model;
a text to speech conversion subsystem that is configured to convert the generated narrative text story to an audio file; and
a narrative audio generator subsystem that is configured to generate a narrative audio by combining the audio file from the text to speech conversion subsystem with the generated emotional arc, wherein the narrative audio is modulated with corresponding pitch and tone based on a value of the generated emotional arc; and
a narrative voice output system outputs the generated narrative audio to the patient in a voice of a robot.
|