DETAILED ACTION
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 12/18/2025 has been entered.
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 Arguments
Applicant’s arguments with respect to claim(s) 1, 3-9, 11 – 15, 17 – 20 have been considered but are moot in view of the new ground of rejection. Upon further review, the examiner deems that the claim as amended are ineligible under 101 abstract idea.
The prior art rejection has been withdrawn.
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, 3-9, 11-15, 17-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. For purposes of this rejection, claims 1, 9, and 15 are treated as representative independent claims. Claims 3–8, 11–14, and 17–20 are evaluated as dependent claims. Claims 2, 10, and 16 are canceled.
Claims 1, 3-9, 11-15, 17-20 are rejected because they are directed to the abstract idea of data manipulation and mathematical processing of voice samples using algorithmic perturbation techniques to achieve de-identification relative to a classifier model.
The claims recite the abstract idea of mathematical concepts and data processing, including algorithmic analysis, feature selection, optimization, and transformation of data.
Claim 1 recites, in relevant part:
receiving audio data associated with at least one voice sample;
perturbing one or more voice samples toward an edge of a decision boundary of at least one classifier model based on using a gradient-based perturbation algorithm;
perturbing at least one pitch of each perturbed voice sample after the voice sample is perturbed with the gradient-based perturbation algorithm to shift each perturbed voice sample across the decision boundary of the classifier model using a non-gradient based perturbation algorithm; and
encoding a media file with the de-identified voice sample.
These limitations recite abstract concepts because they amount to: receiving and processing data; applying mathematical algorithms to the data; evaluating and selecting features based on computed impact; optimizing the data to achieve a desired classification result; and outputting modified data.
Claims 3, 11, and 17 recite computing a gradient, determining a direction of the gradient, and injecting a perturbation based on the gradient and direction. These are mathematical operations and algorithmic manipulations of data.
Claims 4, 12, and 18 recite a fast gradient signed method (FGSM), which is a known mathematical algorithm.
Claims 5, 13, and 19 recite determining a spectrum of pitches, inputting the spectrum into an algorithm to determine impact, selecting a pitch based on the determined impact, and injecting a perturbation to shift the sample across the decision boundary. These are also analytical and algorithmic steps.
Claims 6, 7, 8, 14, and 20 further narrow the abstract idea using additional algorithmic or result-oriented limitations, but do not alter the fundamental abstract character of the claims.
Accordingly, claims 1, 3-9, 11-15, 17-20 recite an abstract idea. (STEP 2A-1)
The claims do not integrate the abstract idea into a practical application. Although the claims are tied to audio data, voice samples, classifier models, and media files, these elements are merely used as the environment in which the abstract idea is performed. The claims do not recite any particular technological improvement in audio processing, classifier operation, computer functionality, or media encoding.
The claims recite only generic computer components, including:
at least one processor;
at least one non-transitory computer-readable medium;
instructions;
encoding a media file.
These components are recited at a high level of generality and do not impose meaningful limits on the abstract idea beyond implementing it on a computer.
The claims do not recite a specific improvement to the functioning of a computer, a network, an audio codec, a classifier model, or any other technical field.
Instead, the claims merely use a computer to process voice samples, apply perturbation algorithms, and generate de-identified output. The claims are largely framed in terms of the desired outcome, such as: shifting voice samples across a decision boundary, providing de-identified voice samples, and achieving imperceptibility. Such result-oriented language does not amount to a practical application of the abstract idea.
Accordingly, the claims are not integrated into a practical application. (STEP 2A-2)
Individually, the additional elements do not amount to significantly more than the abstract idea itself. The claims recite conventional computing components, such as a processor and computer-readable medium, that perform their ordinary functions of receiving, processing, and storing data. The claims further recite routine operations, including: computing a gradient; determining a direction; injecting a perturbation; selecting a pitch based on impact; and encoding a media file.
These are generic computer-implemented steps and do not amount to an inventive concept. The additional limitations involving: gradient-based perturbation, FGSM, genetic algorithms, evolutionary algorithms, differential evolutionary algorithms, pitch spectrum analysis, and imperceptibility are themselves abstract algorithmic refinements of the same underlying idea and do not add a meaningful limitation that transforms the claim into patent-eligible subject matter.
The claims do not recite an unconventional arrangement of components or a specific technological improvement in the implementation of the algorithms. The ordered combination of limitations merely automates an abstract process using generic computing resources.
Therefore, the claims do not include an inventive concept sufficient to transform the abstract idea into patent-eligible subject matter. (STEP 2-B).
For the reasons set forth above, claims 1, 3-9, 11-15, 17-20 are rejected under 35 U.S.C. § 101 as being directed to an abstract idea without reciting significantly more than the abstract idea itself.
Allowable Subject Matter
Claims 1, 3-9, 11-15, 17-20 allowable over the prior art of record.
The following is a statement of reasons for the indication of allowable subject matter: the prior art of record, taken alone or in combination, fail to teach a method and system for processing data comprising, inter alia, perturbing, with the at least one processor, at least one pitch of each perturbed voice sample of the at least one voice sample after the at least one voice sample is perturbed with the gradient-based perturbation algorithm to shift each perturbed voice sample of the at least one voice sample across the decision boundary of the at least one classifier model to provide at least one de-identified voice sample based on using a non-gradient based perturbation algorithm.
The 101 abstract idea rejection can be overcome if the independent claims are amended to recite for example,
Claim 1 after line 6, insert:
determining, with the at least one processor, a spectrum of pitches from the at least one voice sample using a pitch tracking algorithm;
inputting the spectrum of pitches into a non-gradient based perturbation algorithm comprising a differential evolution algorithm to determine a level of impact of perturbing each pitch;
…
and
encoding, with the at least one processor, a media file with the at least one de-identified voice sample.
Claim 1, line 12, after de-identified voice sample, insert: wherein a difference between the at least one voice sample and the at least one de-identified voice sample is imperceptible to a human listener; and
The claims will therefore recite a particular signal processing pipeline solving a specific technical problem, as identified in the specification, and improving the technical quality of the audio transformation.
The above languages are presented only as a suggestion to the applicant based on the specification and as understood by the examiner.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Deng et al., “Decision-Based Attack to Speaker Recognition System via Local Low-Frequency Perturbation”, Jan. 01, 2022.
Deng et al., discloses a query-efficient decision-based attack against speaker recognition systems using local low-frequency perturbation. It searches for a local attacking region and then applies perturbations in the low-frequency domain to reduce distortion and improve attack success. The method is designed to work in a black-box setting where only predicted labels are available. It teaches improved attack success rate, lower query cost, and lower distortion compared with prior methods. But fails to teach the claim 1 particular feature of perturbing one pitch of each voice sample using a non-gradient-based perturbation algorithm after a gradient-based perturbation step. It also does not disclose selecting pitch from a spectrum of pitches based on highest impact or the specific pitch-based de-identification workflow recited in the claims.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to RICHEMOND DORVIL whose telephone number is (571)272-7602. The examiner can normally be reached 8:30 - 5:30 M-F.
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.
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.
/RICHEMOND DORVIL/ Supervisory Patent Examiner, Art Unit 2658
Prosecution Insights