DETAILED ACTION
This Office Action is in response to the communication filed on 12/23/2025.
Claims 1-2, 4-10 and 12-14 are pending.
Claims 1, 4-5, 9-10 and 12-14 have been amended.
Claims 3 and 11 have been canceled.
Claims 1-2, 4-10 and 12-14 are rejected.
The Examiner cites particular sections in the references as applied to the claims below for the convenience of the applicant(s). Although the specified citations are representative of the teachings in the art and are applied to the specific limitations within the individual claim, other passages and figures may apply as well. It is respectfully requested that, in preparing responses, the applicant(s) fully consider the references in their entirety as potentially teaching all or part of the claimed invention, as well as the context of the passage as taught by the prior art or disclosed by the Examiner.
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 .
Notes
Claim 4 is listed as “original” however claim 4 has been amended.
Specification
The objection to the disclosure has been withdrawn.
Claim Rejections - 35 USC § 101
The 101 rejection has been withdrawn due to updated guidance, specifically changes to the MPEP in light of Ex Parte Desjardins.
Claim Rejections - 35 USC § 112
The 112 rejections have been withdrawn due to Applicant’s amendments.
Response to Arguments
Applicant's arguments filed 12/23/2025 have been fully considered but they are not persuasive.
Applicant argues that Cheng is limited to temporal segmentation of speech and speech-to-text recognition, and Cheon is directed to approximate operations on numerical data that neither reference recognizes or suggests a text-based privacy-preserving communication architecture in which unstructured text data is vectorized, homomorphically encrypted, and then transmitted to an external server for query/database searching, as presented in the claims. Examiner disagrees.
Chang does perform speech to text conversions by segmenting unstructured data and analyzing it. Cheon does perform mathematical operations including homomorphically encrypting which provides privacy-preserving communications for outsourcing computations. The rejection is made by incorporating the privacy-preserving teachings of Cheon into the segmenting and analyzing unstructured data of Chang. One cannot show nonobviousness by attacking references individually where the rejections are based on combinations of references. See In re Keller, 642 F.2d 413, 208 USPQ 871 (CCPA 1981); In re Merck & Co., 800 F.2d 1091, 231 USPQ 375 (Fed. Cir. 1986). Therefore Chang in view of Cheon does arrive at the invention as disclosed and interpreted under BRI of claim 1.
In response to applicant's argument that the examiner's conclusion of obviousness is based upon improper hindsight reasoning, it must be recognized that any judgment on obviousness is in a sense necessarily a reconstruction based upon hindsight reasoning. But so long as it takes into account only knowledge which was within the level of ordinary skill at the time the claimed invention was made, and does not include knowledge gleaned only from the applicant's disclosure, such a reconstruction is proper. See In re McLaughlin, 443 F.2d 1392, 170 USPQ 209 (CCPA 1971).
Applicant claims the if a POSITA applied the Homomorphic encryption scheme of Cheon to the pre-processor of Cheng then the neural network would fail because it cannot process homomorphically encrypted ciphertexts. However, a POSITA would understand that Homomorphic encryption allows for the processing of encrypted data without needing to decrypt it first, including speech-to-text systems, because homomorphic encryption simply provides a way to perform computations on encrypted data, including speech data, ensuring that the output is equivalent to what would be produced from the original data when decrypted, thus preserving privacy and security. Therefore applying the homomorphic encryption scheme of Cheon to the pre-processor of Cheng would not cause downstream failures but would instead increase downstream security.
Applicant argues that Cheng does not disclose “index” but does disclose “order”. Index is interpreted according to the specifications where index can be an order (instant [0089] the information (e.g., index information related to order (e.g., word order, sentence order, voice order, etc.), index information related to location, attribute information for attributes of unstructured data) related to the unstructured data may have values in an imaginary number). Therefore Cheng at least suggests as index similar to the index in the claimed invention.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1-2 and 6-7, 9-10 and 13-14 are rejected under 35 U.S.C. 103 as being unpatentable over Cheng (U.S. 20210295845), in view of Cheon (U.S. 20200036511).
Regarding claims 1, 9 and 14,
Cheng discloses: A method of processing an encrypted message in an electronic apparatus, the method comprising:
dividing text data into sentence units; (Cheng [0011, 0037-0041] The speech audio may also be divided into multiple alternate data chunks that each represent an alternate chunk of multiple alternate chunks of the speech audio… an indication of each string of blank symbols from the CTC output that is at least as long as the predetermined blank threshold length as a likely sentence pause of a second candidate set of likely sentence pauses)
calculating a vector value of a predetermined size corresponding to each sentence by using a predetermined encoding algorithm for each sentence unit; (Cheng [0011, 0037-0041]; [0179- 0189] teaches determining (calculating) vector(s) of a pre-determined based on a matrix of weights (predetermined encoding algorithm), which are determined during training, for each sentence unit)
…a sequence index for each of a plurality of sentences in the text data is generated, the generated sequence index is (Cheng [Fig. 15B and 15C object 2131 and [00215-0223] teaches a sequence index [0216-217] More specifically, and starting with the data chunk 2131a that represents the temporally earliest chunk of the speech audio of the speech data set 2130, a window 2236 that covers a preselected quantity of temporally consecutive ones of the data chunks 2131a may be shifted across the length of the speech audio, starting with the temporally earliest data chunk 2131a and proceeding throughout all of the data chunks 2131a in temporal order toward the temporally last data chunk 2131a)
Cheng does not explicitly disclose: generating a homomorphic encrypted message by performing homomorphic encryption on the calculated vector value, and
…wherein, in the generating of the homomorphic encrypted message… an encrypted vector value corresponding to the sequence index are inserted into a respective slot in the homomorphic encrypted message.
transmitting the homomorphic encrypted message using a communication device of the electronic apparatus,
However, in the same field of endeavor Cheon discloses: generating a homomorphic encrypted message by performing homomorphic encryption on the calculated vector value, and (Cheon [0032]; [0038-0052] the homomorphic encrypted message generated by the first and second devices 100-1 and 100-2 may be generated in the form that a result value including a message and an error value is restored when decrypted using a secret key later. Hereinafter, in this specification, the message including an error will be referred to as an approximate message.
…wherein, in the generating of the homomorphic encrypted message, (Cheon [Fig. 4 and Fig. 5]; [0084-0095]; teaches generating the homomorphically encrypted message)
an encrypted vector value corresponding to the sequence index are inserted into a respective slot in the homomorphic encrypted message. (Cheon [Fig. 4 and Fig. 5]; [0084-0095]; The homomorphic encrypted message 10, 20 include approximate message regions 11, 21, respectively. The approximate message regions 11 and 21 include messages and errors m1+e1 and m2+e2) together, respectively; this teaches inserting respective values into respective slots in the homomorphic encryption process)
transmitting the homomorphic encrypted message using a communication device of the electronic apparatus (Cheon [Fig. 1-100, 0038-0041] teaches a device homomorphically encrypts data and transmitted the homomorphically encrypted data)
Cheng and Cheon are analogous art because they are from the same field of endeavor communication.
Before the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art, having the teachings of Cheng and Cheon before him or her, to modify the method of Cheng to include the homomorphic encryption of Cheon because it will maintain security and prevent resource and time being wasted in the process of decrypting the encrypted data when a message encrypted by the encryption technique is transmitted to a counterpart.
The motivation for doing so would be [“ to maintain security and improve of communication when a message encrypted by the encryption technique is transmitted to a counterpart”] (Paragraph [0003-0004] by Cheon)].
Therefore, it would have been obvious to combine Cheng and Cheon to obtain the invention as specified in the instant claim.
Regarding claims 2 and 10,
Cheng in view of Cheon discloses: The method as claimed in claim 1, wherein,
for each sentence unit, and (Cheng [0011, 0037-0041]; [0179- 0189] teaches determining (calculating) vector(s) of a pre-determined based on a matrix of weights (predetermined encoding algorithm), which are determined during training, for each sentence unit)
Cheon further discloses: in the generating of the homomorphic encrypted message, homomorphic encryption is performed on each vector value generated (Cheon [0032]; [0038-0052] the homomorphic encrypted message generated by the first and second devices 100-1 and 100-2 may be generated in the form that a result value including a message and an error value is restored when decrypted using a secret key later. Hereinafter, in this specification, the message including an error will be referred to as an approximate message)
each homomorphic encrypted vector value is sequentially put into a plurality of slots in the homomorphic encrypted message to generate the homomorphic encrypted message. (Cheon [Fig. 4 and Fig. 5]; [0084-0095]; The homomorphic encrypted message 10, 20 include approximate message regions 11, 21, respectively. The approximate message regions 11 and 21 include messages and errors m1+e1 and m2+e2) together, respectively.
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify with Cheon for similar reasons as cited in claim 1.
Regarding claim 6,
Cheng in view of Cheon discloses: The method as claimed in claim 1, further comprising: converting voice data into the text data, wherein, in the dividing of the text data into the sentence units, the converted text data is divided into text units. (Cheng [0002-0015; 0026-0037; 0265] It has become commonplace to perform automated speech-to-text conversion of captured speech audio; in temporal order, to form text data that represents the text into which the speech audio of the speech data set has been converted (e.g., the text data 2519))
Regarding claim 7,
Cheng in view of Cheon discloses: The method as claimed in claim 1, wherein the text data is at least one of a text message and a chat message. (Cheng [0002; 0086; 0142; 0162; 0196; 0229; 0232] a conversion to text may be performed as part of indexing and/or memorializing the contents of recorded voice messages or of phone conversations)
Regarding claim 13,
Cheng in view of Cheon discloses: The arithmetic unit as claimed in claim 9, wherein, when voice data is input, the processor converts the input voice data into text data and stores the text data in the memory. (Cheng [0002-0015; 0026-0037; 0117; 0140; 0265] It has become commonplace to perform automated speech-to-text conversion of captured speech audio; in temporal order, to form text data that represents the text into which the speech audio of the speech data set has been converted (e.g., the text data 2519))
Claims 4 and 12 is rejected under 35 U.S.C. 103 as being unpatentable over Cheng (U.S. 20210295845), in view of Cheon (U.S. 20200036511) in further view of Polyakov (U.S. 20220376891).
Regarding claims 4 and 12,
Cheng in view of Cheon discloses: The method as claimed in claim 1, wherein,
Cheon further discloses: in the generating of the homomorphic encrypted message, the homomorphic encrypted message is generated by (Cheon [Fig. 4 and Fig. 5]; [0084-0095]; teaches generating the homomorphically encrypted message) placing the encrypted numerical data in a real number area of the homomorphic encrypted message and (Cheon [0087] it is possible to perform a real number operation in an encrypted state, and efficiency of encryption/decryption may be increased) placing the encrypted (Cheon [Fig. 4 and Fig. 5]; [0084-0095]; The homomorphic encrypted message 10, 20 include approximate message regions) sequence index in (Cheng [Fig. 15B and 15C object 2131 and [00215-0223] teaches a sequence index)
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify with Cheon for similar reasons as cited in claim 1.
Cheng in view of Cheon does not explicitly disclose: an imaginary number area of the homomorphic encrypted message.
However, in the same field of endeavor Polyakov discloses: an imaginary number area of the homomorphic encrypted message. (Polyakov [0002-0019]; [0041-0051] FIG. 2A and FIG. 2B are a flowchart for a method 200 of cryptography, according to some embodiments. Method 200 may be a method of cryptography based on 128 bit integers that precludes reducing the 128 bit integer from being segmented into smaller words before encryption. Method 200 may include receiving… The input may be an n-dimensional vector with numeric entries which have a real part and an imaginary part)
Cheng in view of Cheon and Polyakov are analogous art because they are from the same field of endeavor word approximation and homomorphic encryption.
Before the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art, having the teachings of Cheng in view of Cheon and Polyakov before him or her, to modify the method of Cheng in view of Cheon to include the real and imaginary inputs of Polyakov because it will allow for improved security using large-word approximate homomorphic encryption.
The motivation for doing so would be [“Advantages of the invention may include mitigating IND-CPA+attacks by adding random noise and/or rounding result during decryption. Advantages of the invention may be due to using a noise magnitude (or e.g., equivalently rounding precision) that is significantly higher than currently known approximation noise, by at least, for example, 20 or 30 bits. This may result in the RNS moduli needing to be increased by e.g., 20-30 bits, requiring residue arithmetic with numbers larger than 64 bits.”] (Paragraph [0006-0007] by Polyakov)].
Therefore, it would have been obvious to combine Cheng in view of Cheon and Polyakov to obtain the invention as specified in the instant claim.
Claim 5 is rejected under 35 U.S.C. 103 as being unpatentable over Cheng (U.S. 20210295845), in view of Cheon (U.S. 20200036511) in further view of Kuperman (U.S. 20170244737).
Regarding claim 5,
Cheng in view of Cheon discloses: The method as claimed in claim 1, wherein the vector value of the predetermined size is represented as (Cheng [0011, 0037-0041]; [0179- 0189] teaches determining (calculating) vector(s) of a pre-determined based on a matrix of weights (predetermined encoding algorithm), which are determined during training, for each sentence unit)
While Cheng in view of Chron discloses bit values, Cheng in view of Chron does not explicitly disclose:
However, in the same field of endeavor Kuperman discloses: a 32-bit real value within a range of [-1, 1]. (Kuperman [0074] the attribute may be represented as an attribute vector of size 1×1... the model generator 209 may consider the values of combined (and 1×1) feature vectors for classification. In some embodiments, integer value attributes values are normalized to a value range of 0 to 1 or alternatively −1 to 1)
Cheng in view of Cheon and Kuperman are analogous art because they are from the same field of endeavor vectorization for machine learning.
Before the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art, having the teachings of Cheng in view of Cheon and Kuperman before him or her, to modify the method of Cheng in view of Cheon to include the vector normalization of Kuperman.
The motivation for doing so would be [“to avoid excessive weight to any attribute represented by real numbers that may potentially have large values”] (Paragraph [0074] by Kuperman)].
Therefore, it would have been obvious to combine Cheng in view of Cheon and Kuperman to obtain the invention as specified in the instant claim.
Claim 8 is rejected under 35 U.S.C. 103 as being unpatentable over Cheng (U.S. 20210295845), in view of Cheon (U.S. 20200036511), in view of Polyakov (U.S. 20220376891) and in further view of Pathak (U.S. 20220108085).
Regarding claim 8,
Cheng in view of Cheon discloses: The method as claimed in claim 4, wherein
Cheng in view of Cheon does not explicitly disclose: the predetermined encoding algorithm is a bidirectional encoder representations from transformers (BERT) language model.
However, in the same field of endeavor Pathak discloses: the predetermined encoding algorithm is a bidirectional encoder representations from transformers (BERT) language model. (the NLP language model may be used in combination with the reference indicator during training, these models may include but are not limited to ELMO and BERT. These modules generally utilize one or more neural networks. Similarly, other NLP techniques associated with a more general linguistic approach may be used. The scope of such is beyond this application)
Cheng in view of Cheon and Pathak are analogous art because they are from the same field of endeavor Naturel language processing.
Before the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art, having the teachings of Cheng in view of Cheon and Pathak before him or her, to modify the method of Cheng in view of Cheon to include the BERT language model of Pathak because it will Allow for language to be processed by neural networks, specifically speech to text.
The motivation for doing so would be [”The technology disclosed in the present application is directed to automating the entire (or at least part) of the above aforesaid procedure. As will be shown, in later sections, this may be achieved by exploiting advances made in the area of NLP, deep convolutional neural networks using images (e.g., image captioning)”, “These modules generally utilize one or more neural network”] (Paragraph [0039-0055] by Pathak)].
Therefore, it would have been obvious to combine Cheng in view of Cheon and Pathak to obtain the invention as specified in the instant claim.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's
disclosure.
Gentry 6/18/2019 (US 20200403781) teaches homomorphic encryption.
Seo 8/30/2021 (US 20220405474) classification of encrypted data using neural network, and more particularly, to a method, a computing device and a computer-readable medium for classification of encrypted data using neural network to derive an embedding vector by embedding text data encrypted through an encryption technique, input the embedding vector to a feature extraction module to which a plurality of neural network models are connected, and enable the encrypted text data to be labeled without a separate decryption process by labeling the encrypted text data with a specific classification item based on a learning vector including a feature value derived from the feature extraction module.
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 THOMAS A CARNES whose telephone number is (571)272-4378. The examiner can normally be reached Monday-Friday.
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, Shewaye Gelagay can be reached at (571) 272-4219. 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.
THOMAS A. CARNES
Examiner
Art Unit 2436
/THOMAS A CARNES/Examiner, Art Unit 2436 /SHEWAYE GELAGAY/Supervisory Patent Examiner, Art Unit 2436