Back to results
US 11,759,145 B1Granted
Technique for identifying dementia based on voice data
AssigneeHAII corp.
InventorsHo Yung Kim, Dong Han Kim, Hye Bin Hwang, Chan Yeong Park, Ji An Choi, Hyun Jeong Ko, Su Yeon Park, Byung Hun Yun
Life of this patent
Estimating the term…
Abstract
Disclosed is a method of identifying dementia by at least one processor of a device. The method may include obtaining content analysis information and voice activity analysis information by using a plurality of first voice data and second voice data which have been obtained by performing a plurality of tasks in a user terminal, and identifying dementia by inputting the content analysis information and the voice activity analysis information to a dementia identification model.
Claims12
A method of identifying, by at least one processor of a device, dementia, the method comprising:
- obtaining content analysis information and voice activity analysis information by using a plurality of first voice data and a second voice data which have been obtained by performing a plurality of tasks in a user terminal;
- and
- identifying dementia by inputting the content analysis information and the voice activity analysis information to a dementia identification model,
- wherein the content analysis information comprises information on a first ratio of at least one keyword stored in storage to all words spoken by a user, which have been generated by analyzing the plurality of first voice data and the second voice data, and information on a second ratio of at least one analogue related to the at least one keyword to all the words,
- wherein the information on the first ratio comprises:
- information on a ratio of an N-th keyword related to an N-th image displayed in the user terminal when N-th voice data, among the plurality of first voice data, is obtained to all words spoken by the user, which have been generated by analyzing the N-th voice data;
- information on a ratio of an (N+1)-th keyword related to an (N+1)-th image displayed in the user terminal when (N+1)-th voice data, among the plurality of first voice data, is obtained to all words spoken by the user, which has been generated by analyzing the (N+1)-th voice data;
- and
- information on a ratio of a plurality of keywords related to a plurality of images comprising the N-th image and the (N+1)-th image displayed in the user terminal when the plurality of first voice data is obtained to all words spoken by the user, which have been generated by analyzing the second voice data,
- wherein the information on the second ratio comprises:
- information on a ratio of an N-th analogue mapped to the N-th keyword related to the N-th image displayed in the user terminal when the N-th voice data, among the plurality of first voice data, is obtained to all the words spoken by the user, which have been generated by analyzing the N-th voice data;
- information on a ratio of an (N+1)-th analogue mapped to the (N+1)-th keyword related to the (N+1)-th image displayed in the user terminal when the (N+1)-th voice data, among the plurality of first voice data, is obtained to all the words spoken by the user, which have been generated by analyzing the (N+1)-th voice data;
- and
- information on a ratio of a plurality of analogues mapped to the plurality of keywords related to the plurality of images comprising the N-th image and the (N+1)-th image displayed in the user terminal when the plurality of first voice data is obtained to all the words spoken by the user, which have been generated by analyzing the second voice data,
- wherein the (N+1)-th image is an image changed in response to a touch input when the N-th image is displayed, and
- wherein the N is a natural number equal to or greater than 1.
13 more elements
A computer program stored in a computer-readable storage medium, wherein when the computer program is executed by at least one processor of a device, the computer program performs steps of identifying dementia, wherein the steps comprises:
- obtaining content analysis information and voice activity analysis information by using a plurality of first voice data and second voice data which have been obtained by performing a plurality of tasks;
- and
- identifying dementia by inputting the content analysis information and the voice activity analysis information to a dementia identification model,
- wherein the content analysis information comprises information on a first ratio of at least one keyword stored in storage to all words spoken by a user, which have been generated by analyzing the plurality of first voice data and the second voice data, and information on a second ratio of at least one analogue related to the at least one keyword to all the words,
- wherein the information on the first ratio comprises:
- information on a ratio of an N-th keyword related to an N-th image displayed in the user terminal when N-th voice data, among the plurality of first voice data, is obtained to all words spoken by the user, which have been generated by analyzing the N-th voice data;
- information on a ratio of an (N+1)-th keyword related to an (N+1)-th image displayed in the user terminal when (N+1)-th voice data, among the plurality of first voice data, is obtained to all words spoken by the user, which has been generated by analyzing the (N+1)-th voice data;
- and
- information on a ratio of a plurality of keywords related to a plurality of images comprising the N-th image and the (N+1)-th image displayed in the user terminal when the plurality of first voice data is obtained to all words spoken by the user, which have been generated by analyzing the second voice data,
- wherein the information on the second ratio comprises:
- information on a ratio of an N-th analogue mapped to the N-th keyword related to the N-th image displayed in the user terminal when the N-th voice data, among the plurality of first voice data, is obtained to all the words spoken by the user, which have been generated by analyzing the N-th voice data;
- information on a ratio of an (N+1)-th analogue mapped to the (N+1)-th keyword related to the (N+1)-th image displayed in the user terminal when the (N+1)-th voice data, among the plurality of first voice data, is obtained to all the words spoken by the user, which have been generated by analyzing the (N+1)-th voice data;
- and
- information on a ratio of a plurality of analogues mapped to the plurality of keywords related to the plurality of images comprising the N-th image and the (N+1)-th image displayed in the user terminal when the plurality of first voice data is obtained to all the words spoken by the user, which have been generated by analyzing the second voice data,
- wherein the (N+1)-th image is an image changed in response to a touch input when the N-th image is displayed, and
- wherein the N is a natural number equal to or greater than 1.
13 more elements
A device for identifying dementia, the device comprising:
- storage in which at least one program instruction has been stored;
- and
- at least one processor configured to perform the at least one program instruction,
- wherein the at least one processor is configured to:
- obtain content analysis information and voice activity analysis information by using a plurality of first voice data and second voice data which have been obtained by performing a plurality of tasks;
- and
- identify dementia by inputting the content analysis information and the voice activity analysis information to a dementia identification model,
- wherein the content analysis information comprises information on a first ratio of at least one keyword stored in storage to all words spoken by a user, which have been generated by analyzing the plurality of first voice data and the second voice data, and information on a second ratio of at least one analogue related to the at least one keyword to all the words,
- wherein the information on the first ratio comprises:
- information on a ratio of an N-th keyword related to an N-th image displayed in the user terminal when N-th voice data, among the plurality of first voice data, is obtained to all words spoken by the user, which have been generated by analyzing the N-th voice data;
- information on a ratio of an (N+1)-th keyword related to an (N+1)-th image displayed in the user terminal when (N+1)-th voice data, among the plurality of first voice data, is obtained to all words spoken by the user, which has been generated by analyzing the (N+1)-th voice data;
- and
- information on a ratio of a plurality of keywords related to a plurality of images comprising the N-th image and the (N+1)-th image displayed in the user terminal when the plurality of first voice data is obtained to all words spoken by the user, which have been generated by analyzing the second voice data,
- wherein the information on the second ratio comprises:
- information on a ratio of an N-th analogue mapped to the N-th keyword related to the N-th image displayed in the user terminal when the N-th voice data, among the plurality of first voice data, is obtained to all the words spoken by the user, which have been generated by analyzing the N-th voice data;
- information on a ratio of an (N+1)-th analogue mapped to the (N+1)-th keyword related to the (N+1)-th image displayed in the user terminal when the (N+1)-th voice data, among the plurality of first voice data, is obtained to all the words spoken by the user, which have been generated by analyzing the (N+1)-th voice data;
- and
- information on a ratio of a plurality of analogues mapped to the plurality of keywords related to the plurality of images comprising the N-th image and the (N+1)-th image displayed in the user terminal when the plurality of first voice data is obtained to all the words spoken by the user, which have been generated by analyzing the second voice data,
- wherein the (N+1)-th image is an image changed in response to a touch input when the N-th image is displayed, and
- wherein the N is a natural number equal to or greater than 1.
17 more elements
Dependencies are inferred from the claim text.