Granted Patent B2

US 12,726,574 · App. 18/599,866 · Granted Sep 1, 2026

Inventors: Mansi Goyal (Cupertino, CA); Ajoy K Singh (Cupertino, CA); Andrey Berezin (Cupertino, CA)

Assignee: Apple Inc.

H04M3/53333G10L15/22

Abstract

Claims (47)

1 . A method comprising:

receiving, by a user device, a voicemail directed to a user associated with the user device;

analyzing the voicemail, by an artificial intelligence agent, to determine an action to perform on behalf of the user based on contents of the voicemail, wherein the artificial intelligence agent also utilizes user device context information and a collection of past voicemails to analyze the voicemail to determine the action to perform on behalf of the user based on the contents of the voicemail;

outputting, by the artificial intelligence agent, a task that is relevant to the voicemail and that should be performed on behalf of the user; and

based on the outputting of the task that is relevant to the voicemail, automatically performing the task that is relevant to the voicemail on behalf of the user.

2 . The method of claim 1 , wherein the user device context information includes one or more of a location of the user device, applications in use on the user device, an application in focus on the user device, and status of a limited interruption setting.

3 . The method of claim 1 , further comprising:

prior to or contemporaneous with analyzing the voicemail, determining, by an urgency determination engine, to classify the voicemail as urgent; and

when the voicemail is determined to be urgent, interrupting the user, by the urgency determination engine, even if the user device has a limited interruption setting enabled.

4 . The method of claim 3 , wherein the artificial intelligence agent and the urgency determination engine are part of a same machine learning model.

5 . The method of claim 1 , wherein the artificial intelligence agent also outputs a task confidence score along with the task, and wherein the task confidence score indicates a confidence by the artificial intelligence agent that the task should be performed on behalf of the user.

6 . The method of claim 3 , further comprising:

determining, with the urgency determination engine, whether the voicemail has frequency characteristics indicative of a non-human voice or periods of silence that are indicative of a non-human voice; and

classifying the voicemail as not urgent if the urgency determination engine determines the voicemail has the frequency characteristics indicative of a non-human voice or the periods of silence that are indicative of a non-human voice.

7 . The method of claim 3 , wherein automatically performing the task that is relevant to the voicemail on behalf of the user is performed in response to the urgency determination engine determining that the voicemail is not urgent.

8 . A system comprising:

one or more processors; and

at least one computer-readable storage medium having stored therein instructions which, when executed by the one or more processors, cause the one or more processors to perform operations comprising:

receiving, by a user device, a voicemail directed to a user associated with the user device;

analyzing the voicemail, by an artificial intelligence agent, to determine an action to perform on behalf of the user based on contents of the voicemail, wherein the artificial intelligence agent also utilizes user device context information and a collection of past voicemails to analyze the voicemail to determine the action to perform on behalf of the user based on the contents of the voicemail;

outputting, by the artificial intelligence agent, a task that is relevant to the voicemail and that should be performed on behalf of the user; and

based on the outputting of the task that is relevant to the voicemail, automatically performing the task that is relevant to the voicemail on behalf of the user.

9 . The system of claim 8 , wherein the user device context information includes one or more of a location of the user device, applications in use on the user device, an application in focus on the user device, and status of a limited interruption setting.

10 . The system of claim 8 , wherein the operations further comprise:

prior to or contemporaneous with analyzing the voicemail, determining, by an urgency determination engine, to classify the voicemail as urgent; and

when the voicemail is determined to be urgent, interrupting the user, by the urgency determination engine, even if the user device has a limited interruption setting enabled.

11 . The system of claim 10 , wherein the artificial intelligence agent and the urgency determination engine are part of a same machine learning model.

12 . The system of claim 8 , wherein the artificial intelligence agent also outputs a task confidence score along with the task, and wherein the task confidence score indicates a confidence by the artificial intelligence agent that the task should be performed on behalf of the user.

13 . The system of claim 10 , wherein the operations further comprise:

determining, with the urgency determination engine, whether the voicemail has frequency characteristics indicative of a non-human voice or periods of silence that are indicative of a non-human voice; and

classifying the voicemail as not urgent if the urgency determination engine determines the voicemail has the frequency characteristics indicative of a non-human voice or the periods of silence that are indicative of a non-human voice.

14 . The system of claim 10 , wherein automatically performing the task that is relevant to the voicemail on behalf of the user is performed in response to the urgency determination engine determining that the voicemail is not urgent.

15 . A non-transitory computer-readable storage medium having stored therein instructions which, when executed by one or more processors, cause the one or more processors to perform operations comprising:

receiving, by a user device, a voicemail directed to a user associated with the user device;

analyzing the voicemail, by an artificial intelligence agent, to determine an action to perform on behalf of the user based on contents of the voicemail, wherein the artificial intelligence agent also utilizes user device context information and a collection of past voicemails to analyze the voicemail to determine the action to perform on behalf of the user based on the contents of the voicemail;

outputting, by the artificial intelligence agent, a task that is relevant to the voicemail and that should be performed on behalf of the user; and

based on the outputting of the task that is relevant to the voicemail, automatically performing the task that is relevant to the voicemail on behalf of the user.

16 . The non-transitory computer-readable storage medium of claim 15 , wherein the user device context information includes one or more of a location of the user device, applications in use on the user device, an application in focus on the user device, and status of a limited interruption setting.

17 . The non-transitory computer-readable storage medium of claim 15 , wherein the operations further comprise:

prior to or contemporaneous with analyzing the voicemail, determining, by an urgency determination engine, to classify the voicemail as urgent; and

when the voicemail is determined to be urgent, interrupting the user, by the urgency determination engine, even if the user device has a limited interruption setting enabled.

18 . The non-transitory computer-readable storage medium of claim 15 , wherein the artificial intelligence agent also outputs a task confidence score along with the task, and wherein the task confidence score indicates a confidence by the artificial intelligence agent that the task should be performed on behalf of the user.

19 . The non-transitory computer-readable storage medium of claim 17 , wherein the operations further comprise:

determining, with the urgency determination engine, whether the voicemail has frequency characteristics indicative of a non-human voice or periods of silence that are indicative of a non-human voice; and

classifying the voicemail as not urgent if the urgency determination engine determines the voicemail has the frequency characteristics indicative of a non-human voice or the periods of silence that are indicative of a non-human voice.

20 . The non-transitory computer-readable storage medium of claim 17 , wherein automatically performing the task that is relevant to the voicemail on behalf of the user is performed in response to the urgency determination engine determining that the voicemail is not urgent.

21 . The non-transitory computer-readable storage medium of claim 17 , wherein the artificial intelligence agent and the urgency determination engine are part of a same machine learning model.

Assignments (1)

Continuity (1)

References Cited (12)