Prosecution Insights
Last updated: April 19, 2026
Application No. 18/374,455

SYSTEMS AND METHODS FOR ADDRESSING POSSIBLE INTERRUPTION DURING INTERACTION WITH DIGITAL ASSISTANT

Non-Final OA §103
Filed
Sep 28, 2023
Examiner
SHAH, PARAS D
Art Unit
2653
Tech Center
2600 — Communications
Assignee
Adeia Guides Inc.
OA Round
3 (Non-Final)
74%
Grant Probability
Favorable
3-4
OA Rounds
3y 9m
To Grant
99%
With Interview

Examiner Intelligence

Grants 74% — above average
74%
Career Allow Rate
474 granted / 645 resolved
+11.5% vs TC avg
Strong +31% interview lift
Without
With
+31.1%
Interview Lift
resolved cases with interview
Typical timeline
3y 9m
Avg Prosecution
24 currently pending
Career history
669
Total Applications
across all art units

Statute-Specific Performance

§101
20.3%
-19.7% vs TC avg
§103
44.9%
+4.9% vs TC avg
§102
13.8%
-26.2% vs TC avg
§112
10.5%
-29.5% vs TC avg
Black line = Tech Center average estimate • Based on career data from 645 resolved cases

Office Action

§103
DETAILED ACTION 1. This communication is in response to the Amendments and Arguments (RCE) filed on 2/25/2026. Claims 33-35, 37-45, 47-52 are pending and have been examined. Claims 1-32, 36, 46 are cancelled. Response to Amendments and Arguments 2. With respect to claim rejections under 35 USC 101 (mental process), the applicant’s amendments and arguments are carefully considered and the rejections are withdrawn based on the amended limitation “a learning model trained based at least in part on a gradient descent data classification” which specifically recites a learning model trained by a particular training process. Applicant's arguments with respect to claim rejections under 35 USC 103 have been fully considered, but they are not persuasive. In particular, the applicant specifically argues that for the independent claims, the cited references do not teach “wherein the data of the other digital assistant sessions is maintained in a database and classified using a learning model trained based at least in part on a gradient descent data classification ..” In response, the examiner respectfully disagrees. Note that CHEN teaches: [Summary, para 3] “Voice input .. input to the recognition system .. in the BP neural network training stage, mainly analyze the features .. build a model for each entry, and save it as a template library <read on ‘database’>” and Mavrovouniotis teaches: [Abstract] “an ant colony optimization (ACO) algorithm is applied to train feed-forward neural networks for pattern classification <where ‘pattern’ is subject to BRI such as a topic> .. In addition, the ACO training algorithm is hybridized with gradient descent training.” The applicant further argues that the references do not teach “accessing the database to identify one or more entries related to the topic, wherein the one or more entries indicate that prior notifications of the particular type were typically snoozed during the other digital assistant sessions by delaying the prior notifications until the completion of the other digital assistant sessions ..” In response, the examiner respectfully disagrees. Note that CHEN teaches: [Summary, para 3] “Voice input .. input to the recognition system .. in the BP neural network training stage, mainly analyze the features .. build a model for each entry, and save it as a template library <read on ‘database’>” and NASIR teaches: [0033] “creates alerts to the user based on learnings from previous projects <read on ‘prior notifications during the other digital assistant sessions’>).” POLLACK teaches: [Abstract] “Autominder achieves this goal by providing adaptive, personalized reminders of .. activities of daily living .. make decisions about whether and when it is most appropriate to issue reminders” which reads on “prior notifications of the particular type were typically snoozed .. automatically causing the notification to be delayed until after the digital assistant session is complete ..” Claim Rejections - 35 USC § 103 3. Claims 33-35, 37-45, 47-52 are rejected under 35 U.S.C. 103 as being unpatentable over Nasir (US 20210224753; hereinafter NASIR) in view of Chen, et al. (CN109979436A; hereinafter CHEN), further in view of Pollack, et al. (Elsevier, 2003; hereinafter POLLACK) and further in view of Mavrovouniotis, et al. (Soft Comput 2015; hereinafter Mavrovouniotis). As per claim 33, NASIR (Title: Artificial intelligence enabled scheduler and planner) discloses “A method comprising: detecting an initiation of a digital assistant session based at least in part [ receiving a voice command; based at least in part on receiving the voice command, processing the voice command using natural language processing ] to determine a topic (NASIR, Fig. 2 - request inputs of a project .. schedule meetings of the project <read on ‘initiation of a digital assistant session’ where project reads on ‘topic’>); computing a predicted length of time of the digital assistant session based at least in part on [ comparing the digital assistant session to data of other digital assistant sessions relating to the topic, wherein the data of the other digital assistant sessions is maintained in a database ] associated with a user profile, and wherein the data of the other digital assistant sessions is pre-processed and [ classified using a learning model trained based at least in part on a gradient descent data classification ] (NASIR, [0008], FIG. 3 .. determining which time slots are flexible in order to schedule a meeting <read on ‘a predicted length of time’> between required parties; [0005], resolving schedule planning conflicts .. based on user preferences, user role importance, and user time commitments <read on ‘user profile’ for stored data>); [ determining that a notification of a particular type is scheduled to occur during the predicted length of time of the digital assistant session; accessing the database to identify one or more entries related to the topic, wherein the one or more entries indicate that prior notifications of the particular type were typically snoozed during the other digital assistant sessions by delaying the prior notifications until the completion of the other digital assistant sessions based at least in part on determining the respective occurrence of the prior notifications during the other digital assistant session; and based at least in part on identifying the one or more entries of the database indicating that the prior notifications of the particular type were typically snoozed during the other digital assistant sessions, automatically causing the notification to be delayed until after the digital assistant session is complete ] (NASIR, [0033], planning program 200 readjusts other task <read on ‘topic’> schedules of the template of the project based on dependencies and creates alerts to the user based on learnings from previous projects <read on ‘prior notifications during the other digital assistant sessions’>).” NASIR does not explicitly disclose “receiving a voice command; based at least in part on receiving the voice command, processing the voice command using natural language processing .. comparing the digital assistant session to data of other digital assistant sessions relating to the topic, wherein the data of the other digital assistant sessions is maintained in a database ..” However, the limitation is taught by CHEN (Title: A kind of BP neural network speech recognition system and method based on frequency spectrum adaptive method). In the related field of endeavor, CHEN teaches: [Summary, para 3] “Voice input .. input to the recognition system .. in the BP neural network training stage, mainly analyze the features .. build a model for each entry, and save it as a template library <read on ‘database’>. In the recognition stage .. a test template is generated, it is matched with the reference template, and the recognition result is generated <read on ‘natural language processing’ and ‘comparing the digital assistant session to data of other digital assistant sessions relating to the topic’ for any subsequent action such as determining the predicted session length based on the topic>.” Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate the teachings of CHEN in the system (as taught by NASIR) for the purpose of identifying digital session topics based on voice commands so as to determine predicted session length for session scheduling. NASIR in view of CHEN does not explicitly disclose “determining that a notification of a particular type is scheduled to occur during the predicted length of time of the digital assistant session; (accessing the database to identify one or more entries related to the topic) (see CHEN above), wherein the one or more entries indicate that prior notifications of the particular type were typically snoozed during the other digital assistant sessions by delaying the prior notifications until the completion of the other digital assistant sessions based at least in part on determining the respective occurrence of the prior notifications during the other digital assistant session; and based at least in part on identifying the one or more entries of the database indicating that the prior notifications of the particular type were typically snoozed during the other digital assistant sessions, automatically causing the notification to be delayed until after the digital assistant session is complete.” However, the limitation is taught by POLLACK (Title: Autominder: an intelligent cognitive orthotic system for people with memory impairment). In the related field of endeavor, POLLACK teaches: [Abstract] “Autominder achieves this goal by providing adaptive, personalized reminders of .. activities of daily living <where activities read on topics/entries, and the associated types for notification> … Autominder uses a range of AI techniques to model an individual’s daily plans, observe and reason about the execution of those plans, and make decisions about whether and when it is most appropriate to issue reminders” where “an individual’s daily plans” reads on “user profile” and “observe and reason about the execution of those plans” reads on employing various factors and conditions to determine the process of sending notifications, and “whether and when it is most appropriate to issue reminders” reads on “a notification of a particular type is scheduled to occur .. prior notifications of the particular type were typically snoozed .. automatically causing the notification to be delayed until after the digital assistant session is complete” which includes snooze/delay operations. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate the teachings of POLLACK in the system (as taught by NASIR and CHEN) based on using various factors and conditions to automatically determine whether and when to issue what type of notifications. NASIR in view of CHEN and POLLACK does not explicitly disclose “classified using a learning model trained based at least in part on a gradient descent data classification ..” However, the limitation is taught by Mavrovouniotis (Title: Training neural networks with ant colony optimization algorithms for pattern classification). In the related field of endeavor, Mavrovouniotis teaches: [Abstract] “an ant colony optimization (ACO) algorithm is applied to train feed-forward neural networks for pattern classification <where ‘pattern’ is subject to BRI such as a topic> .. In addition, the ACO training algorithm is hybridized with gradient descent training.” Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate the teachings of Mavrovouniotis in the system (as taught by NASIR, CHEN and POLLACK) for data/topic classification based on a gradient descent training for digital assistant session applications such as for session scheduling. As per claim 34 (dependent on claim 33), NASIR in view of CHEN, POLLACK and Mavrovouniotis further discloses “wherein the digital assistant session is a voice session between a user associated with the user profile and the digital assistant (NASIR, Fig. 3; [0016], Network 110 .. capable of receiving and transmitting data, voice, and/or video signals, including multimedia signals that include voice, data, and video information <read on a ready mechanism to support meeting sessions involving any multimedia signals, as typically involving voice/audio, image/video and text in a meeting>).” As per claim 35 (dependent on claim 33), NASIR in view of CHEN, POLLACK and Mavrovouniotis further discloses “wherein the data of the other digital assistant sessions comprises a history of previous digital assistant sessions involving a user associated with the user profile and the digital assistant relating to the topic, and a history of previous digital assistant sessions involving other users and other digital assistants relating to the topic (NASIR, Fig. 3; [0033], planning program 200 readjusts other task <read on ‘topic’> schedules of the template of the project based on dependencies and creates alerts to the user <read on ‘notification’> based on learnings from previous projects <read on ‘previous sessions .. other users ..’> where a similar delay impacts a risk of timely delivery of the project).” As per claim 37 (dependent on claim 33), NASIR in view of CHEN, POLLACK and Mavrovouniotis further discloses “based at least in part on the determining that the notification of the particular type is scheduled to occur during the predicted length of time of the digital assistant session: generating for output details relating to the occurrence of the notification, wherein the details include at least one of a subject matter of the notification and a time that the occurrence of the notification is scheduled to be presented (NASIR, [0033], If planning program 200 determines that a time slot of a meeting of a project is not satisfactory .. planning program .. creates alerts to the user <read on ‘notification .. output details .. subject matter .. time’>).” As per claim 38 (dependent on claim 33), NASIR in view of CHEN, POLLACK and Mavrovouniotis further discloses “wherein the notification type includes one of: a notification relating to a user's calendar, a notification relating to a user's email account, or a notification relating to a user's smart device (NASIR, [0019], a calendar application that planning program 200 utilizes to retrieve available time slots for a user; [0033], planning program 200 readjusts other task schedules of the template of the project based on dependencies and creates alerts to the user <read on ‘notification’> based on learnings from previous projects where a similar delay impacts a risk of timely delivery of the project).” As per claim 39 (dependent on claim 33), NASIR in view of CHEN, POLLACK and Mavrovouniotis further discloses “receiving user information related to a user associated with the user profile responding to the prior notifications of the particular type that occurred during the other digital assistant sessions with the digital assistant (NASIR, [0005], resolving schedule planning conflicts .. based on user preferences, user role importance, and user time commitments <read on ‘user information .. user profile’>; [0033], planning program 200 readjusts other task schedules of the template of the project based on dependencies and creates alerts to the user <read on ‘notification’> based on learnings from previous projects <read on ‘the other digital assistant sessions’ and ‘responding to the prior notifications’> where a similar delay impacts a risk of timely delivery of the project).” As per claim 40 (dependent on claim 33), NASIR in view of CHEN, POLLACK and Mavrovouniotis further discloses “wherein the determining that the notification of the particular type is scheduled to occur during the predicted length of time of the digital assistant session is based at least in part on scheduling data associated with a user associated with the user profile (NASIR, [0005], resolving schedule planning conflicts <where conflicts read on ‘a notification of a particular type is scheduled to occur’ which is subject to BRI> .. based on user preferences, user role importance, and user time commitments <read on ‘user profile’>).” As per claim 41 (dependent on claim 33), NASIR in view of CHEN, POLLACK and Mavrovouniotis further discloses “based at least in part on accessing the database to identify the one or more entries indicating that prior notifications of the particular type were not typically snoozed, providing a prompt comprising selectable options to delay the digital assistant session until after the occurrence of the notification is complete or to delay or cancel the occurrence of the notification (See claim 33. NASIR, [0033], If planning program 200 determines that a time slot of a meeting of a project is not satisfactory .. planning program 200 readjusts other task schedules of the template of the project based on dependencies and creates alerts to the user based on learnings from previous projects; POLLACK, [Abstract], Autominder .. make decisions about whether and when it is most appropriate to issue reminders <read on prior notifications of the particular type were typically snoozed or not and other options> .. automatically causing the notification to be delayed until after the digital assistant session is complete).” As per claim 42 (dependent on claim 33), NASIR in view of CHEN, POLLACK and Mavrovouniotis further discloses “wherein the automatically causing the notification to be delayed is performed without generating for an output a notification to a user associated with the user profile (See Claim 33).” Claims 43-45, 47-52 (similar in scope to claims 33-35, 37-42) are rejected under the same rationale as detailed above for claims 33-42. Claim 43 further discloses “receive a user interface input .. the I/O circuitry is configured to output the notification after the delay (NASIR, Fig. 3 and Fig. 2 - request inputs of a project .. schedule meetings of the project; [0033], If planning program 200 determines that a time slot of a meeting of a project is not satisfactory .. planning program .. creates alerts to the user <read on ‘notifications’>. Also see claim 33 for more details).” Conclusion 4. Any inquiry concerning this communication or earlier communications from the examiner should be directed to FENG-TZER TZENG whose telephone number is 571-272-4609. The examiner can normally be reached on M-F (9:00-5:00). The fax phone number where this application or proceeding is assigned is 571-273-4609. If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Paras Shah (SPE) can be reached on 571-270-1650. Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000. /FENG-TZER TZENG/ 3/5/2026 Primary Examiner, Art Unit 2653
Read full office action

Prosecution Timeline

Sep 28, 2023
Application Filed
Jun 14, 2025
Non-Final Rejection — §103
Sep 12, 2025
Response Filed
Dec 02, 2025
Final Rejection — §103
Feb 12, 2026
Examiner Interview Summary
Feb 12, 2026
Applicant Interview (Telephonic)
Feb 25, 2026
Request for Continued Examination
Mar 02, 2026
Response after Non-Final Action
Mar 07, 2026
Non-Final Rejection — §103 (current)

Precedent Cases

Applications granted by this same examiner with similar technology

Patent 12586591
SOUND SIGNAL DECODING METHOD, SOUND SIGNAL DECODER, PROGRAM, AND RECORDING MEDIUM
2y 5m to grant Granted Mar 24, 2026
Patent 12579367
TWO-TOWER NEURAL NETWORK FOR CONTENT-AUDIENCE RELATIONSHIP PREDICTION
2y 5m to grant Granted Mar 17, 2026
Patent 12579360
LEARNING SUPPORT APPARATUS FOR CREATING MULTIPLE-CHOICE QUIZ
2y 5m to grant Granted Mar 17, 2026
Patent 12562173
WEARABLE DEVICE CONTROL BASED ON VOICE COMMAND OF VERIFIED USER
2y 5m to grant Granted Feb 24, 2026
Patent 12559026
VEHICLE AND CONTROL METHOD THEREOF
2y 5m to grant Granted Feb 24, 2026
Study what changed to get past this examiner. Based on 5 most recent grants.

AI Strategy Recommendation

Get an AI-powered prosecution strategy using examiner precedents, rejection analysis, and claim mapping.
Powered by AI — typically takes 5-10 seconds

Prosecution Projections

3-4
Expected OA Rounds
74%
Grant Probability
99%
With Interview (+31.1%)
3y 9m
Median Time to Grant
High
PTA Risk
Based on 645 resolved cases by this examiner. Grant probability derived from career allow rate.

Sign in with your work email

Enter your email to receive a magic link. No password needed.

Personal email addresses (Gmail, Yahoo, etc.) are not accepted.

Free tier: 3 strategy analyses per month