Granted Patent B2

US 12,645,734 · App. 18/975,878 · Granted Jun 2, 2026

Inventors: Olivia Nauman (Phoenix, AZ); Lauren Mitchell (Phoenix, AZ); James Rodriguez (Goodyear, AZ); Ivan Gutierrez (Laveen, AZ); Anish Agarwal (Bloomington, IL); Ruben Tequida (Phoenix, AZ)

Assignee: State Farm Mutual Automobile Insurance Company

G06F16/639G06N5/022

Abstract

Claims (66)

1 . An auditory biometric system for outputting a playlist file comprising at least one audio file, the auditory biometric system comprising at least one processor, and at least one memory in communication with the at least one processor, the at least one processor programmed to:

receive user biometric data detected by a sensor associated with a user computer device, wherein user biometric data is associated with a biometric state of a user;

apply user biometric data to a trained auditory biometric model to generate model outputs comprising the playlist file, wherein the auditory biometric model is trained using training data including a plurality of historical records associated with a plurality of different historical users, wherein the historical records each include a historical audio file and historical user biometric data of a user collected while the historical audio file was being played; and

transmit a playlist message including the playlist file to the user computer device for execution by the user computer device.

2 . The auditory biometric system of claim 1 , wherein the at least one processor is further programmed to:

receive a user selected input from the user computer device, wherein the user selected input includes at least one of a target biometric state, a playlist profile, an activity, or a duration of activity; and

apply the user selected input and the user biometric data to the trained auditory biometric model to generate the playlist file.

3 . The auditory biometric system of claim 1 , wherein the user biometric data includes one or more of heart rate, heart rate variability, body or skin temperature, skin conductivity, skin moisture or sweat, breathing rate, blood oxygen levels, step rate/number, motion data, acceleration, speed, Global Positioning System data, blood pressure, VO 2 max, calories burned, step count, sleep/awake detection, chronotypes, circadian rhythm, electric heart rate or electrocardiogram.

4 . The auditory biometric system of claim 1 , wherein the at least one processor is further programmed to:

build a first training dataset including a plurality of historical records associated with a plurality of different historical users, wherein the plurality of historical records each include a historical audio file and historic user biometric data of a user collected while the historical audio file was being played; and

train, in a first session, an auditory biometric model using the first training dataset to generate the trained auditory biometric model.

5 . The auditory biometric system of claim 4 , wherein the at least one processor is further programmed to:

build a second training dataset including a first plurality of historical records associated with a plurality of different historical users and a second plurality of historical records associated with a specific current user, wherein the historical records each include a historical audio file and historical user biometric data of a user collected while the historical audio file was being played; and

train, in a second session, an auditory biometric model using the second training dataset to generate the trained auditory biometric model.

6 . The auditory biometric system of claim 1 , wherein the at least one processor is further programmed to:

build a training dataset including a plurality of historical records associated with a specific current user, wherein the historical records each include a historical audio file and historical user biometric data of the specific current user collected while the historical audio file was being played; and

re-train an auditory biometric model using the training dataset to generate the trained auditory biometric model.

7 . The auditory biometric system of claim 1 , wherein the at least one processor is further programmed to:

receive updated user biometric data detected by the sensor associated with the user computer device, wherein updated user biometric data is associated with a current biometric state of a user;

apply the updated user biometric data to the trained auditory biometric model to generate model outputs including an updated playlist file, the updated playlist file including at least one song different from the at least one song of the previously generated playlist file; and

transmit an updated playlist message including the updated playlist file to the user computer device.

8 . The auditory biometric system of claim 1 , wherein the at least one processor is further programmed to:

build a training dataset including a plurality of historical records associated with a plurality of different historical users and a plurality of historical records associated with a specific current user, wherein the historical records each include a historical audio file and historical user biometric data of a user collected while the historical audio file was being played, and wherein the plurality of historical records associated with the specific current user includes weighting factors that are greater than weighting factors for historical records associated with a plurality of different historical users; and

train an auditory biometric model using the training dataset to generate the trained auditory biometric model.

9 . The auditory biometric system of claim 8 , wherein the historical records each include an audio record and historical user biometric data associated with the audio record, wherein the audio record includes at least one song and at least one metric of the song, wherein the at least one metric includes at least one of: genre, tempo, tempo range, rhythm, melody, timbre, and harmony.

10 . The auditory biometric system of claim 1 , wherein the at least one processor is further programmed to:

receive updated user biometric data detected by the sensor associated with the user computer device, wherein updated user biometric data is associated with a current heart rate of a user;

receive a user selected input from the user computer device, wherein the user selected input includes at least one of a target biometric state, the target biometric state including a target heart rate;

apply the updated user biometric data and the target heart rate to the trained auditory biometric model to generate model outputs including an updated playlist file, the updated playlist file including at least one song that is different from the at least one song of the previously generated playlist file, wherein the at least one different song has an increased tempo compared to the at least one song of the previously generated playlist file; and

transmit an updated playlist file message including the updated playlist file to the user computer device.

11 . The auditory biometric system of claim 1 , wherein the at least one processor is further programmed to:

apply a time of day comprising a morning period to the trained auditory biometric model to generate model outputs including an updated playlist file, the updated playlist file including at least one different song that is different from the at least one song of the previously generated playlist file, wherein the at least one different song has an increased tempo compared to the at least one song of the playlist file; and

transmit an updated playlist message including the updated playlist file to the user computer device.

12 . The auditory biometric system of claim 1 , wherein the at least one processor is further programmed to:

receive first updated user biometric data detected by the sensor associated with the user computer device, wherein the first updated user biometric data is associated with a heart rate of the user at a first time;

receive a user selected input from the user computer device, wherein the user selected input includes at least one of a target biometric state, the target biometric state including a target heart rate;

apply the first updated user biometric data and the target heart rate to the trained auditory biometric model to generate model outputs including a first updated playlist file, the first updated playlist file including at least one different song that is different from the at least one song of the previously generated playlist file, wherein the at least one different song has an increased tempo compared to the at least one song of the previously generated playlist file; and

transmit a first updated playlist message including the first updated playlist file to the user computer device;

receive second updated user biometric data detected by the sensor associated with the user computer device, wherein the second updated user biometric data is associated with a heart rate of the user at a second time after the first time;

apply the second updated user biometric data and the target heart rate to the trained auditory biometric model to generate model outputs including a second updated playlist file, the second updated playlist file including at least one different song that is different from the at least one song of the first updated playlist file, wherein the at least one different song has an increased tempo compared to the at least one song of the first updated playlist file; and

transmit a second updated playlist message including the second updated playlist file to the user computer device.

13 . The auditory biometric system of claim 1 , wherein the at least one processor is further programmed to:

receive user data detected by the sensor associated with a user computer device, wherein the user data comprises surrounding noise data detected by a microphone of the user computer device;

apply user data to a trained auditory biometric model to generate model outputs comprising the playlist file and an audio level; and

transmit a playlist message including the playlist file to the user computer device and an audio level, wherein the playlist message includes instructions that when executed by the user computer device cause the user computer device to adjust a volume of a speaker to the determined audio level.

14 . A computer-implemented auditory biometric method for outputting a playlist file comprising at least one audio file, the auditory biometric method implemented using a computer device including at least one processor in communication with a memory device, the method comprising, via the computer device and/or at least one processor:

receiving user biometric data detected by a sensor associated with a user computer device, wherein user biometric data is associated with a biometric state of a user;

applying user biometric data to a trained auditory biometric model to generate model outputs comprising the playlist file, wherein the auditory biometric model is trained using training data including a plurality of historical records associated with a plurality of different historical users, wherein the historical records each include a historical audio file and historical user biometric data of a user collected while the historical audio file was being played; and

transmitting a playlist message including the playlist file to the user computer device for execution by the user computer device.

15 . The method of claim 14 , wherein the method further includes:

receiving a user selected input from the user computer device, wherein the user selected input includes at least one of a target biometric state, a playlist profile, an activity, or a duration of activity; and

applying the user selected input and the user biometric data to the trained auditory biometric model to generate the playlist file.

16 . At least one non-transitory computer-readable storage medium including computer-executable instructions embodied thereon for outputting a playlist file comprising at least one audio file, wherein when executed by at least one processor, the computer-executable instructions cause the processor to:

receive user biometric data detected by a sensor associated with a user computer device, wherein user biometric data is associated with a biometric state of a user;

apply user biometric data to a trained auditory biometric model to generate model outputs comprising the playlist file, wherein the auditory biometric model is trained using training data including a plurality of historical records associated with a plurality of different historical users, wherein the historical records each include a historical audio file and historical user biometric data of a user collected while the historical audio file was being played; and

transmit a playlist message including the playlist file to the user computer device for execution by the user computer device.

17 . The at least one non-transitory computer-readable storage medium of claim 16 , wherein the computer-executable instructions cause the processor to:

receive a user selected input from the user computer device, wherein the user selected input includes at least one of a target biometric state, a playlist profile, an activity, or a duration of activity; and

apply the user selected input and the user biometric data to the trained auditory biometric model to generate the playlist file.

18 . The at least one non-transitory computer-readable storage medium of claim 16 , wherein the user biometric data includes one or more of heart rate, heart rate variability, body or skin temperature, skin conductivity, skin moisture or sweat, breathing rate, blood oxygen levels, step rate/number, motion data, acceleration, speed, Global Positioning System data, blood pressure, VO 2 max, calories burned, step count, sleep/awake detection, chronotypes, circadian rhythm, electric heart rate or electrocardiogram.

19 . The at least one non-transitory computer-readable storage medium of claim 16 , wherein the computer-executable instructions cause the processor to:

build a first training dataset including a plurality of historical records associated with a plurality of different historical users, wherein the plurality of historical records each include a historical audio file and historical user biometric data of a user collected while the historical audio file was being played; and

train, in a first session, an auditory biometric model using the first training dataset to generate the trained auditory biometric model.

20 . The at least one non-transitory computer-readable storage medium of claim 16 , wherein the computer-executable instructions cause the processor to:

build a second training dataset including a first plurality of historical records associated with a plurality of different historical users and a second plurality of historical records associated with a specific current user, wherein the historical records each include a historical audio file and historical user biometric data of a user collected while the historical audio file was being played; and

train, in a second session, an auditory biometric model using the second training dataset to generate the trained auditory biometric model.

Assignments (1)

Continuity (2)

References Cited (11)