Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Status of Claims
Claims 1-3, and 9-17 have been examined in this Final Rejection and are pending.
Claims 4-8, and 18-20 were previously canceled.
Priority
Application 17/844,011 filed 06/19/2022 claims priority to provisional application 63/212,700 filed 06/20/2021.
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claim(s) 1-3, and 9-17 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention.
Lack of Written Description/Not in Specification
Claim 1 recites:
"a recurrent neural network (RNN) comprising a hidden state memory",
"modifying, during runtime, the hidden state memory based on one or more streaming parameters including buffer occupancy or text arrival rate"
"selectively persisting a subset of prior hidden states identified as context-critical while discarding non-critical hidden states to maintain a bounded state size"
"generating output text prior to receipt of an end-of-sequence indicator"
Claim 15 recites:
a plurality of cloud-based edge nodes configured to receive partial text segments from at least one source in real time;
a stream segmentation module executing on each edge node, the module configured to generate variable-length text segments based on semantic boundary detection;
a distributed ordering controller configured to dynamically assign, based on measured inter-node latency, a sequence-assignment role to a selected edge node
a reconciliation engine configured to merge text segments received from multiple edge nodes using a conflict-free replicated data structure:
a client interface configured to stream the merged text segments in sequence as they are generated
As best understood, the portions of the disclosure most closely related to the above indicated recitation are in the specification as filed at 0076, 0087, 0099, 00117, 0068, and 0047.
0076 teaches supervised learning algorithms include but not limited to the AI/ML models, NLP+ algorithms, and outlier detection programs to detect changes in patterns and translate them into actionable recommendations and that AI/ML models may be but not limited to artificial neural networks and support vector machines. The Specification fails to teach a recurrent neural network (RNN), hidden state memory, or parameters including buffer occupancy or text arrival rate.
00122 teaches analyzing comments data for specific keywords and generating a list of usernames related to the specific keywords. The Specification fails to teach generating output text prior to receipt of an end-of-sequence indicator.
0087 teaches that the software takes the comments data from the read module, processes it as per the required engagement tool, and renders it as a visualization that the presenter 209 can share with the audience to engage them by showing a visual representation of their inputs in real time. The Specification fails to teach a plurality of cloud-based edge nodes configured to receive partial text segments from at least one source in real time.
0099 teaches Fig. 12 illustrating a keyword analysis module that enables the StreamAlive 209 software to use NLP to analyze the comments data to identify keywords 1203. Also, an NLP model is created to analyze short form text 1205. The Specification fails to teach a stream segmentation module executing on each edge node, the module configured to generate variable-length text segments based on semantic boundary detection.
00117 teaches the StreamAlive 209 software analyzing the comment data for relevance, frequency of interaction, etc. for each participant, in real time 1403, in other embodiments more parameters are used for analysis. Using sentiment analysis, a score will be assigned for each interaction of each audience member 1405. Then an overall score will be assigned to the audience member 1407 during the livestream. The Specification fails to teach a distributed ordering controller configured to dynamically assign, based on measured inter-node latency, a sequence-assignment role to a selected edge node.
0068 teaches a recommendation engine 615, analytics engine for convert module 617 and features engine 613. The Specification fails to teach a reconciliation engine configured to merge text segments received from multiple edge nodes using a conflict-free replicated data structure.
0047 teaches the audience interacting with the presenter 209 via the chat interface provided by the online meeting / livestreaming software platform 213. The StreamAlive application 207 further captures the chat output of the audience and provides meaningful feedback to the presenter 209 who in turn can better engage with the audience 211. The Specification fails to teach a client interface configured to stream the merged text segments in sequence as they are generated.
Accordingly, support in the disclosure is not found for Claim 1, and Claim 15. Dependent Claims 2-3, 9-14, and 16-17 are rejected due to their dependencies on Claim 1, and Claim 15.
Subject Matter Distinguishable from Prior Art
Claims 1-3, and 9-17 would be allowable if rewritten or amended to overcome the rejection(s) under 35 U.S.C. § 112 set forth in this Office Action.
As previously disclosed in the Non-Final Rejection on 12/26/2024, Claims 1-3, and 9-17 are subject matter distinguishable from prior art.
Response to Arguments
Applicant's arguments filed 067 with respect to 35 U.S.C. § 112, have been fully considered but they are not persuasive. Applicant argues that a person of ordinary skill in the art would have understood that neural-network architectures configured to process sequential data streams, including recurrent neural networks, LSTM networks, GRU networks, sequence-processing networks, and related architectures, maintain internal state information that is updated as successive inputs are received. The specification discloses “artificial neural networks and support vector machines” in [0076], however, is silent on recurrent neural networks comprising hidden-state memory. The standard is not what is obvious to one of ordinary skill in the art, but what is disclosed in the original specification. Therefore, this argument is not persuasive.
The Applicant further argues that the recited runtime modification of hidden-state memory based on streaming parameters merely reflects operation of a neural-network engagement model in a real-time livestream environment in which audience comments continuously arrive and are processed as a stream. Applicants augments of retaining selected hidden states while discarding less relevant hidden states is representing one example of managing contextual information within a neural-network model operating on an extended sequence of audience interactions. This is not persuasive because the Examiner cannot find where the hidden state, buffer occupancy or text arrival rate, and discarding non-critical hidden states to maintain a bounded state size is described in the specification. Therefore, this argument is not persuasive
The Applicant further argues that the originally filed specification repeatedly emphasizes real-time feedback, real-time recommendations, and real-time engagement analysis, therefore, the recitation of generating output text prior to receipt of an end-of-sequence indicator similarly reflects incremental output generation during processing of a continuously arriving livestream comment stream since the incremental output generation is a known implantation detail consistent with the disclosed real-time engagement framework. The Examiner agrees that real-time feedback, recommendations, and engagement is disclosed, however, the Examiner cannot locate where generating output text prior to receipt of an end-of-sequence indicator is located in the specification. Therefore, this argument is not persuasive.
The Applicant further argues that the specification discloses a server or cloud-based software-as-a-service platform configured to process livestream audience interactions in real time. The Examiner agrees that a cloud-based software is disclosed, however, the Examiner cannot locate where a plurality of cloud-based edge nodes configured to receive partial text segments from at least one source in real time is disclosed. One of ordinary skill in the art would have not understood that such real-time cloud-based processing environments may be implemented using a variety of distributed-computing techniques, including multiple processing nodes, distributed services, cloud-computing resources, edge-computing resources, distributes queues, distributed databases, synchronization mechanisms, and load-balancing techniques. Therefore, this argument is not persuasive.
The Applicant further argues that the recited stream-segmentation module represents one example of processing continuously arriving livestream data, and the specification repeatedly emphasizes real-time acquisition, parsing, analysis, storage, and processing of livestream audience interactions, and a person of ordinary skill in the art would have understood segmentation of incoming data streams, including segmentation based on semantic content, timing information, or message boundaries, as a known implementation techniques for processing large-scale streaming data. The Examiner respectfully disagrees. The standard is not what is obvious to one of ordinary skill in the art, but what is disclosed in the original specification. The specification discloses NLP, but fails to disclose an edge node, and variable-length text segments based on semantic boundary detection.
The Applicant further argues that the distributed ordering controller and reconciliation engine represent implementation details associated with coordinating outputs generated by multiple processing resources and that a person of ordinary skill in the art would have recognized that distributed processing environments commonly employ synchronization, ordering, replication, and conflict-resolution mechanisms to ensure that outputs generated across multiple processing resources may be combined and presented in a consistent manner. The Examiner respectfully disagrees. The specification fails to disclose an edge node, and a distributed ordering controller, and it would not be apparent to one of ordinary skill in the art to have this combination. The standard is not what is obvious to one of ordinary skill in the art, but what is disclosed in the original specification. Therefore, this argument is not persuasive.
The Applicant further argues that the recited conflict-free replicated data structure (CRDT) similarly represents a particular example of a conflict resolution mechanism that may be employed, therefore, identifying a specific implementation of the broader synchronization and reconciliation functionality that would have been understood by persons of ordinary skill in the art. The Examiner respectfully disagrees. There is no disclosure of a conflict-free data structure (CRDT), therefore this argument is not persuasive.
The Applicant further argues that the recited client interface configured to stream merged text segments in sequence merely reflect presentation of processed outputs to a user in real time, and the originally filed specification repeatedly emphasizes real-time delivery of processed audience information, engagement recommendations, notifications, visualizations, and feedback. The Examiner respectfully disagrees. The specification does not disclose a client interface to stream the merged text segments in sequence as they are generated. Therefore, this argument is not persuasive.
Conclusion
All claims are identical to or patentably indistinct from, or have unity of invention with claims in the application prior to the entry of the submission under 37 CFR 1.114 (that is, restriction (including a lack of unity of invention) would not be proper) and all claims could have been finally rejected on the grounds and art of record in the next Office action if they had been entered in the application prior to entry under 37 CFR 1.114. Accordingly, THIS ACTION IS MADE FINAL even though it is a first action after the filing of a request for continued examination and the submission under 37 CFR 1.114. See MPEP § 706.07(b). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Emily M Kraisinger whose telephone number is (703)756-4583. The examiner can normally be reached M-F 7:30 AM -4:30 PM.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Jessica Lemieux can be reached at 571-270-3445. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/E.M.K./Examiner, Art Unit 3626 /DENNIS W RUHL/Primary Examiner, Art Unit 3626
