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 .
Response to Amendment
Applicant's amendment filed on May 4, 2026 has been entered. Claims 1 – 15 and 17 – 20 have been amended. No claims have been canceled. Claim 21 has been added. Claims 1 – 21 are still pending in this application, with claims 1, 8 and 15 being independent.
Response to Amendment
Regarding rejection under 35 USC § 101
The Applicant alleges: “
Step 2A, Prong One - The Claims Do Not Recite a Mental Process
The Office asserts that the limitation involving generating a summary based on multiple portions of text could be performed mentally and therefore constitutes a mental process. Office Action, at 2-3.
However, this characterization does not apply to the elements recited in the amended claims. Amended claim 1, for example, recites that first information of a portion of text and second information specifying a desired format are provided as inputs to neural networks, and that the neural networks generate summaries based on these inputs. Neural networks are machine-learning models implemented through computational operations performed by
specialized computing systems. Such models process structured inputs using learned parameters and model architectures to generate outputs.
Under the 2019 Revised Patent Subject Matter Eligibility Guidance, a claim recites a mental process only when the claimed steps can practically be performed in the human mind or with pen and paper. See MPEP § 2106.04(a)(2). The claimed elements involving providing structured inputs to neural networks and causing those neural networks to generate summaries based on multiple categories of input information require machine-learning computation and therefore cannot reasonably be performed mentally.
Accordingly, the claims do not recite a mental process and therefore do not recite a judicial exception under Step 2A, Prong One.
Examiner’s response:
The Examiner respectfully disagrees.
The claimed limitations can be interpreted as analyzing data, organizing information, and producing output information. Thus, it can be categorized as mental processes, mathematical concept, or methods of organizing human activities.
The neural networks are additional elements. They are further analyzed in Step 2A, Prong Two, and Step 2B.
The Applicant further alleges: “
Step 2A, Prong Two - The Claims Integrate Any Alleged Abstract Idea Into a Practical Application
Even, assuming arguendo, if the claims could be interpreted as reciting an abstract idea related to summarization, the claims clearly integrate any such concept into a practical application. Amended claim 1, for example, recites a specific technological configuration that defines a particular machine-learning workflow in which contextual information from prior text portions and formatting constraints are provided as structured inputs to neural networks to control summary generation. This configuration specifies how a computing system processes information to produce structured summaries.
The 2019 Guidance explains that claims integrate an abstract idea into a practical application when they apply the alleged exception in a manner that improves computer functionality or otherwise imposes meaningful technological limitations. See MPEP §§ 2106.04(d) and 2106.05(a).
Here, the claims do not merely recite the concept of summarizing information. Instead, the claims recite a specific technical mechanism for controlling neural-network-based summarization using multiple categories of input information, including format-specification inputs that guide the structure of the generated summaries. This constitutes a specific technological implementation rather than a generalized instruction to summarize text.
Accordingly, the claims integrate any alleged abstract idea into a practical application and therefore are not directed to a judicial exception.”
Examiner’s response:
The Examiner again respectfully disagrees.
Neural networks are basically mathematical algorithms. Inputting information to specify output format can be categorized as mental processes, mathematical concept, or methods of organizing human activities.
The claimed limitations merely input and output from neural networks. It is considered as “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f).
The Applicant still further alleges: “
Step 2B - The Claims Recite Significantly More Than Any Alleged Abstract Idea
Even if the claims were considered to recite an abstract idea, which Applicant does not concede, the claims include additional elements that amount to significantly more than the alleged judicial exception.
The claims recite circuitry to provide multiple categories of structured information to neural networks and cause those neural networks to generate summaries based on those inputs. The neural networks operate together with the circuitry as an ordered combination to implement a specific machine-learning architecture for generating summaries.
The Office characterizes the processor, memory, and neural networks as generic components performing basic computer functions. However, the claims do not merely recite the use of generic computing components to perform summarization. Instead, the claims recite a particular configuration of circuitry and neural networks in which multiple types of input information are provided to the neural networks to control summary generation.
When considered as an ordered combination, these elements define a specific technological arrangement for machine-learning based text processing, which is more than a generic instruction to implement an abstract idea on a computer.
Under MPEP § 2106.05(d), an ordered combination of elements may amount to significantly more when the elements operate together in a non-conventional arrangement to perform a technological function. The claimed configuration of circuitry providing contextual and format inputs to neural networks to generate structured summaries constitutes such a technological arrangement.”
Examiner’s response:
Independent claims are so broad. They do not claim any detail for implementing a specific machine-learning architecture for generating summaries.
The additional elements (e.g., processor, memory; and one or more neural networks) simply appending well-understood, routine, conventional activities previously known to the industry, specified at a high level of generality, to the judicial exception (WURC) - see MPEP 2106.05(d) and 2106.07(a)III.
Thus, the claims are abstract idea without a significantly more.
Therefore, the 101 rejection is still maintained
Regarding rejection under 35 USC § 102/103
Applicant’s arguments with respect to claim(s) 1 – 21 have been considered but are moot because of the new ground of rejection.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1 – 21 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e. an abstract idea) without significantly more.
Regarding claims 1, 8 and 15:
Step 1:
Claims 1, 8 and 15 are directed towards a process, machine, manufacture or composition of matter which is/are statutory subject matter.
Step 2A:
Prong 1:
Claims 1, 8 and 15 are directed an idea for generating summaries which is an abstract idea.
Consideration of the claimed elements:
Regarding claims 1, 8 and 15:
The claim in the instant application include:
provide, as input to one or more neural networks, first information of one or more first portions of a text and second information specifying a desired format; and
cause the one or more neural networks to generate one or more summaries of a second portion of the text based, at least in part, on the first information and the second information.
Regarding “
provide, as input to one or more neural networks, first information of one or more first portions of a text and second information specifying a desired format; and
cause the one or more neural networks to generate one or more summaries of a second portion of the text based, at least in part, on the first information and the second information”,
it can be interpreted as analyzing data, organizing information, and producing output information. Thus, it can be categorized as mental processes, mathematical concept, or methods of organizing human activities.
Prong 2:
The claims include additional elements of:
processor, memory; and one or more neural networks.
Regarding “processor, memory; and one or more neural networks”, It is considered as “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f).
They are mere instructions to implement an abstract idea uses a computer as a tool to perform an abstract idea.
Moreover, the claim limitations that are not indicative of integration into a practical application.
Thus, the recited generic additional element (e.g., processor, memory; and one or more neural networks) perform no more than their basic computer function. Generic computer-implementation of a method is not a meaningful limitation that alone can amount to significantly more than an abstract idea. Moreover, when viewed as a whole with such additional element considered as an ordered combination, claims modified by adding a computational algorithm, a generic memory and processor are nothing more than a purely conventional computerized implementation of an idea in the general field of computer processing and do not provide significantly more than an abstract idea.
Accordingly, the claims are directed to an idea of itself, and therefore not patent eligible.
Step 2B:
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception such as improvements to another technology or technical field, or other meaningful limitations beyond generally linking the use of the judicial exception to a particular technological environment.
Moreover, the claim language that may be separate from the abstract idea (i.e., additional elements) include computer processors, computer-readable storage media.
The additional element (e.g., processor, memory; and one or more neural networks) perform only basic function, which would be common to every additional element (e.g., processor, memory; and one or more neural networks).
Thus, the recited generic additional elements (e.g., processor, memory; and one or more neural networks) perform no more than their basic computer function. Generic computer-implementation of a method is not a meaningful limitation that alone can amount to significantly more than an abstract idea. Moreover, when viewed as a whole with such additional element considered as an ordered combination, claims modified by adding additional elements (e.g., processor, memory; and one or more neural networks) simply appending well-understood, routine, conventional activities previously known to the industry, specified at a high level of generality, to the judicial exception (WURC) - see MPEP 2106.05(d) and 2106.07(a)III.
Consequently, the identified additional elements taken into consideration individually or in combination fails to amount of significantly more than the abstract idea above.
Regarding claims 2 – 7, 9 – 14 and 16 - 21, the rejection is based on the same rationale described for claims 1, 8 and 15, because the claims include/inherit the same/similar type of problematic limitation(s) as claims 1, 8 and 15, wherein limitations regarding additional aspect for process “generate …”, “combine …”, modify …”, “is …”, “are …”, “split …”, “transcribe …”, “comprise …” and “iteratively prime …”, is/are of sufficient breadth that it would be substantially directed to or reasonably interpreted as a part of the “mental processes” as the abstract idea (similar to claim as stated above). It is noted that further additional limitation is merely generic/conventional computer component/steps to implement the abstract idea, which is, individually or in combination, not sufficient to amount to significantly more than the judicial exception. Therefore, the claimed invention as a whole is directed to an ineligible subject matter.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1 – 8, 10 – 12, 14 – 17 and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Weisz et al. (US Patent Application Publication 2025/0210037), hereinafter referred as Weisz, in view of Maschmeyer et al. (US Patent Application Publication 2024/0311546), hereinafter referred as Maschmeyer.
Regarding claim 15, Weisz discloses a system (Fig. 5) comprising:
one or more processors (Fig. 5, #514); and
a memory (Fig. 5, #524) storing instructions that, if performed by the one or more processors, are to:
provide, as input to one or more neural networks, first information of one or more first portions of a text (Fig. 3, block 301 – 303, [0093 – 0094]); and
cause the one or more neural networks to generate one or more summaries of a second portion of the text based, at least in part, on the first information (Fig. 3, 305, [0098], “generating an ith summary, of the corresponding summaries, that summarizes the ith transcript portion (1<i≤N). The system can generate the ith summary by processing the ith transcript portion and (i-1)th summary as input, using the LLM (305i)”. Because (i-1)th summary is based on (i-1)th portion of the text, thus, generate summaries of ith (first) portion of a text based, at least in part, on (i-1)th (second) portion of the text).
However, Weisz fails to explicitly disclose the system wherein provide, as input to one or more neural networks, second information specifying a desired format; and cause the one or more neural networks to generate one or more summaries based, at least in part, on the second information.
However, in a similar field of endeavor Maschmeyer discloses a system for prompting a large language model (LLM) to generate a revised text passage with formatting (abstract). In addition, Maschmeyer discloses the system provide, as input to one or more neural networks, second information specifying a desired format; and cause the one or more neural networks to generate one or more text based, at least in part, on the second information ([0134 - 0146], user input text-editing instruction for formatting text, and output from LLM (neural network) text with corresponding format).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Weisz, and provide, as input to one or more neural networks, second information specifying a desired format; and cause the one or more neural networks to generate one or more summaries based, at least in part, on the second information. The motivation for doing this is that the text format can be specified to fit the various purpose of usage.
Regarding claim 16 (depends on claim 15), Weisz discloses the system wherein the one or more neural networks are to generate the text based, at least in part, on an audio file ([0093], audio-based file).
Regarding claim 17 (depends on claim 15), Maschmeyer discloses the system wherein the one or more processors are to cause the one or more neural networks to generate the first information in response to one or more prompts that comprise the one or more first portions of the text ([0134 - 0146], user input text-editing instruction for formatting text, and output from LLM (neural network) text with corresponding format).
Regarding claim 20 (depends on claim 15), Weisz discloses the system wherein the one or more processors are to generate an audio transcript from a meeting recording ([0004, 0014, 0062]), and use the one or more neural networks ([0044 – 0045], RNN) to generate a summary of the audio transcript ([0098]).
Regarding claims 1 – 2, they are corresponding to claims 15 – 16, respectively, thus, they are interpreted and rejected for a same reason set forth for claims 15 – 16.
Regarding claim 3 (depends on claim 1), Weisz discloses the one or more processors wherein the circuitry is to generate one or more prompts to cause the one or more neural networks ([0044 – 0045], RNN) to store the first information of (Fig. 5, storage 524) the one or more first portions of the text ([0100]) and use the stored first information to generate the one or more summaries of the second portion of the text ([0098 - 0100]).
Regarding claim 4 (depends on claim 1), Weisz discloses the one or more processors wherein the circuitry is to combine the one or more summaries of the second portion of the text and one or more other summaries of one or more other portions of the text to generate a summary of the text ([0118], “generates an overall summary for the audio-based file based on the generated corresponding summaries”).
Regarding claim 5 (depends on claim 1), Weisz discloses the one or more processors where the circuitry is to modify a transcript of audio to generate the text (Fig. 3, modify of audio to generate the summary text).
Regarding claim 6 (depends on claim 1), Weisz discloses the one or more processors wherein each of the one or more first portions of the text and the second portion of the text is a transcript of a portion of an audio track (Fig. 3, block 303, [0094 – 0097]).
Regarding claim 7 (depends on claim 1), Weisz discloses the processor wherein the specified desired format comprises at least one of the following attributes of a summary: a predetermined length, a thematic organization of the summary, a speaker-by-speaker organization of the summary, and a predetermined number of topics ([0119], “the system causes rendering of the overall summary for the audio-based file in one or more voices based on the one or more speaker tags, in response to a request for a condensed version of the audio-based file.”).
Regarding claims 8 and 10, they are corresponding to claims 15 and 17, respectively, thus, they are interpreted and rejected for a same reason set forth for claims 15 and 17.
Regarding claim 11 (depends on claim 8), Weisz discloses the method further comprising: combining the one or more summaries of the second portion of the text and one or more other summaries of one or more other portions of the text to generate a summary of the text ([0118], “generates an overall summary for the audio-based file based on the generated corresponding summaries”).
However, Weisz fails to explicitly disclose the method wherein generate a summary of the text in a format specified by a prompt.
However, in a similar field of endeavor Maschmeyer discloses a system for prompting a large language model (LLM) to generate a revised text passage with formatting (abstract). In addition, Maschmeyer discloses the system generate a summary of the text in a format specified by a prompt ([0134 - 0146], user input text-editing instruction for formatting text, and output from LLM (neural network) text with corresponding format).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Weisz, and generate a summary of the text in a format specified by a prompt. The motivation for doing this is that the text format can be specified to fit the various purpose of usage.
Regarding claim 12 (depends on claim 8), Weisz discloses the method further comprising:
splitting an audio track into one or more audio track segments (Fig. 3, block 303, [0094 – 0097]);
transcribing each audio track segment of the one or more audio track segments ([0098]); and
combining transcripts of the transcribed audio track segments to generate the text ([0098]).
Regarding claim 14, it is corresponding to claim 7, thus, it is interpreted and rejected for a same reason set forth for claims claim 7.
Claim(s) 13 is/are rejected under 35 U.S.C. 103 as being unpatentable over Weisz in view of Maschmeyer, and in further view of Zhong et al. (WIPO Patent Application Publication WO 2014/085985), hereinafter referred as Zhong.
Regarding claim 13 (depends on claim 8), Weisz discloses the method further comprising: generate one or more transcripts from an audio track (Fig. 3, block 303, [0094 – 0097]).
However, Weisz fails to explicitly disclose the method further comprising modifying the one or more transcripts by performing at least one of: replacing one or more words with corresponding definitions from a dictionary, adding punctuation, applying capitalization, and adding timestamps; and combining the one or more modified transcripts to generate the text.
However, in a similar field of endeavor Zhong discloses a call transcription system (abstract). In addition, Zhong discloses the system wherein modifying the one or more transcripts by performing adding timestamps; and combining the one or more modified transcripts to generate the text (page 7, para. 8, “307. The conversion step 303 converts the input speech signal into an audio file; the speech-to-text step 306 transcribes the audio file formed by the conversion step 303 into a text file; and the tagging step 307 is a text file formed by the transcription of the speech-to-text step 306. The timestamp of the corresponding audio file is added and all text files after the timestamp are sorted according to the timestamp and merged into a call record text file).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Weisz, and modifying the one or more transcripts by performing at least one of: replacing one or more words with corresponding definitions from a dictionary, adding punctuation, applying capitalization, and adding timestamps; and combining the one or more modified transcripts to generate the text. The motivation for doing this is to provide a record of a call content capable of being retrieved according to time, facilitating both parties of a call or other people in retrieving and inquiring about the call content.
Claim(s) 9 and 18 - 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over Weisz in view of Maschmeyer and in further view of Aseniero et al. (US Patent Application Publication 2023/0023037), hereinafter referred as Aseniero.
Regarding claim 9 (depends on claim 8), Weisz discloses the method wherein the one or more neural networks generate the text based, at least in part, on an audio track that records a meeting ([0004, 0014, 0044 – 0045, 0098 - 0100]).
However, Weisz fails to explicitly disclose the method wherein the meeting is a multi-speaker meeting.
However, in a similar field of endeavor Aseniero discloses a system for generating visualizations of recorded meeting data (abstract). In addition, Aseniero discloses wherein the text is generated from a multi-speaker meeting (Fig. 2, [0102 – 0108]).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Weisz, and the meeting is a multi-speaker meeting. The motivation for doing this is that the Application of Weisz can be extended to cover more situations.
Regarding claim 18 (depends on claim 15), Weisz discloses the system wherein the one or more processors are to transcribe an audio track into one or more segments ([0056]).
However, Weisz fails to explicitly disclose the system wherein combine transcripts of different audio track segments to generate the text.
However, in a similar field of endeavor Aseniero discloses a system for generating visualizations of recorded meeting data (abstract). In addition, Aseniero discloses the system wherein combine transcripts of different audio track segments to generate the text (Fig. 2, 288, [0107], combining transcripts of different audio track segments (different persons, and different timelines) to generate the text).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Weisz, and combine transcripts of different audio track segments to generate the text. The motivation for doing this is that the user can see more text for conversation so that it is more convenient for user to see the whole picture.
Regarding claim 19 (depends on claim 15), Weisz fails to explicitly disclose the system wherein the text comprises timestamps, punctuations, and capitalization.
However, in a similar field of endeavor Aseniero discloses a system for generating visualizations of recorded meeting data (abstract). In addition, Aseniero discloses wherein the text comprises timestamps, punctuations, and capitalization (Fig. 2, [0102 – 0108]).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Weisz, and the text comprises timestamps, punctuations, and capitalization. The motivation for doing this is that the text can cover broader range of categories so that the Application of Weisz can be extended.
Regarding claim 21, no reference is found.
The closest reference Maschmeyer fails to teach the claimed limitations.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). 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 QIAN YANG whose telephone number is (571)270-7239. The examiner can normally be reached on Monday-Thursday 8am-6pm.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Andrew Bee can be reached on 571-270-5183. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300. 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.
/QIAN YANG/
Primary Examiner, Art Unit 2677
Prosecution Insights