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
2. In response to the office action mailed on 12/03/2025, applicant filed an amendment on 02/27/2026. Claims 1, 8, and 15 are amended. Claims 6-7, 13, 19-20, and 23-25 are canceled. Claims 26-28 are new. The pending claims are 1-5, 8-12, 14-18, 21-22, and 26-28.
Response to Arguments
3. Applicant’s arguments with respect to the pending claims have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Claim Rejections - 35 USC § 103
4. 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 (i.e., changing from AIA to pre-AIA ) 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.
Claims 1-5, 8-12, 14-18, 21-22, and 26-28 are rejected under 35 U.S.C. 103 as being unpatentable over Kurian (2018/0174591) in view of Khoury (US 10,553,218), and further in view of Huffman (US 20210256985).
Regarding claim 1, Kurian teaches a computer implemented method comprising:
accessing natural biometric profiles of known speakers in storage, each of the natural biometric profiles including unique biometric characteristics of a corresponding one of the known speakers; (Kurian teaches accessing and storing a plurality of biometric voice profiles including biometric information associated with a plurality of individual speakers. (i.e., unique biometric characteristics of each of the known speakers) Kurian at ¶¶ [0029] - [0038].)
generating, via the at least one generative model, synthetic biometric profiles from the natural biometric profiles based on statistical distributions of biometric characteristics across the natural biometric profiles, collectively, without exposing the unique biometric characteristics of any of the known speakers (Kurian teaches a biometric combinatory device (BCD) collecting synthesized biometric voice profiles from a plurality of speaker voice authentication (SVA) which may be used to generate a synthesized voice for one or more users. Kurian at ¶¶ [0030] - [0038]. (This is representing biometric profiles across the natural biometric profiles, collectively, because across all represented users, all biometric profiles are represented. Therefore, collectively, all profiles are represented without exposing the unique biometric characteristics of any known speakers.)), generating a plurality of random samples from the at least one generative model, and generating the synthetic biometric profiles based upon, at least in part, the plurality of random samples; (Kurian teaches using randomized data from the natural voice profiles to synthesize voice profiles. (i.e., generating a plurality of random sample from the generative model). Kurian at ¶¶ [0029] - [0038].); and
performing, via a biometric verification system, biometric verification on a voice sample of a speaker based on the synthetic biometric profiles. (Kurian teaches using the synthesized biometric voice profiles of the users to authenticate the users. Kurian at Abstract and ¶¶ [0014] - [0016].).
Kurian, however, may not explicitly disclose wherein generating the synthetic biometric profiles includes generating the at least one generative model representative of the natural biometric profiles.
Khoury in the same field of endeavor teaches wherein generating the synthetic biometric profiles include generating includes generating the at least one generative model representative of the natural biometric profiles (Khoury teaches generating one or more similarity scores by comparing the features or embeddings extracted for a particular speaker to features or embeddings of other speakers. (i.e., generating a model). Khoury at ¶ [0210]. Further, Khoury teaches using Gaussian Matrix Models (GMMs) to generate a model embedding (i.e., a generative model which is used to calculate the similarity score.) Khoury at ¶ [0043]. Further, Khoury teaches extracting the embeddings (i.e., model embeddings or generative models) from an audio signal containing any number of speaker utterances for any number of speakers. (i.e., representative of a plurality of natural biometric profiles.) Khoury at ¶¶ [0207] - [0209].)
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date to combine the teachings of Kurian with the teachings of Khoury to provide the limitations of claim 1. Doing so would have improved the speed and operation of the process by increasing the speed of processing as recognized by Khoury at 9:1 – 9:6.
Further, Khoury teaches generating statistical distributions of biometric characteristics via at least one generative model. (Khoury teaches generating a first order Gaussian Mixture Model (GMM) statistics (i.e., generating a generative model) from a plurality of extracted audio features of a universal background model including a plurality of speaker models (i.e., a plurality of natural biometric profiles.) Khoury at 2:62 - 3:9. Therefore, Khoury’s generation of a GMM from a plurality of speaker models in combination with Kurian’s generation of synthetic profiles would be obvious to implement by a simple substitution of Kurian’s natural biometric profiles with Khoury’s Gaussian Mixture Model representing the natural biometric profiles.).
Kurian in view of Khoury may not explicitly disclose wherein the synthetic biometric profiles define a cohort of biometric profiles based on at least one of speaker age, speaker language, or speaker gender.
Huffman in the same field of endeavor teaches wherein the synthetic biometric profiles define a cohort of biometric profiles based on at least one of speaker age, speaker language, or speaker gender ([0084], wherein a characteristic line distinguishes between cohorts of voices based on gender, ethnicity, age, etc.). Therefore, it would have been obvious at the time the application was filed to use the above feature of Huffman with the system of Kurian in view of Khoury, in order to provide generative system that combine both data diversity and privacy protection.
Regarding claim 2, Kurian in view of Khoury (hereinafter Kurian-Khoury) teaches all the limitations of claim 1 as laid out above. Further, Khoury teaches the computer-implemented method of claim 1, wherein each of the natural biometric profiles includes a vector of biometric information associated with a corresponding one of the known speakers. (Khoury teaches a voiceprint (i.e., biometric information associated with an individual) Y is equal to R * X. where R is a random matrix (i.e., a multidimensional vector) and X is a sparse vector. Khoury at 8:37 - 8:67. As such, the result of R * X will also be a vector thus making Y a vector and because Y is also voiceprint, the vector is a voiceprint. Voiceprints are biometric information, therefore the biometric profiles of user information (i.e., natural biometric profiles) include vectors of the user's biometrics, corresponding with one of the known speakers.). Therefore, it would have been obvious at the time the application was filed to use Khoury’s vector information feature with the above combined system, in order to protect speakers privacy.
Regarding claim 3, Kurian-Khoury teaches all the limitations of claim 1 as laid out above. Further, Khoury teaches the computer-implemented method of claim 1, wherein generating the at least one generative model includes generating a plurality of generative models for a plurality of biometric characteristics. (Khoury teaches generating first order Gaussian Mixture Model (GMM) statistics. (i.e., a plurality of generative models) for a plurality of voice prints (i.e., a plurality of biometric characteristics) Khoury at 2:62 - 3:9 and 6:38 - 6:48.). Therefore, it would have been obvious at the time the application was filed to use Khoury’s plurality of generative models for a plurality of biometric characteristics with the above combined system, in order to enhance the accuracy, security, and robustness of biometric systems.
Regarding claim 4, Kurian-Khoury teaches all the limitations of claim 1 as laid out above. Further, Khoury teaches the computer-implemented method of claim 1, wherein the at least one generative model is a multivariate Gaussian distribution. (Khoury teaches the generating a Gaussian Mixture Model (GMM) (e.g., multivariate Gaussian distribution.) and using a normal distribution (i.e., a Gaussian distribution). Khoury at 2:62 - 3:9 and 8:37 - 8:67.). Therefore, it would have been obvious at the time the application was filed to use the multivariate Gaussian distribution feature of Khoury with the above combined system, in order to enable complex relationships and correlations between multiple variables during analysis and computation.
Regarding claim 5, Kurian in view of Khoury may not explicitly disclose wherein the plurality of synthetic biometric profiles define a cohort of voiceprints.
Huffman in the same field of endeavor teaches wherein the plurality of synthetic biometric profiles define a cohort of voiceprints( see [0003], [0011], [0033], [0044], wherein Huffman teaches a voice-to-voice conversion system which generates synthetic voice profiles and transforms a source voice into a target voice by extrapolating synthetic frequency components and voice features/characteristics of the target voice from the generated synthetic voice profiles). Therefore, it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date to combine the above feature of Huffman with the system of Kurian in view of Khoury to provide the plurality of synthetic voiceprints define a cohort of voiceprints for a biometric verification system. Doing so would have increased the efficiency and accuracy of the selection of results.
As per claims 8, 9, 10, 12, 14, system claims 8, 9, 10, 12, 14 and method claims 1-5 are related as apparatus and the method of using same, with each claimed element's function corresponding to the claimed method step. Accordingly claims 8, 9, 10, 12, 14 are similarly rejected under the same rationale as applied above with respect to method claims 1-5. Furthermore, Kurian teaches a computing system comprising at least one processor; and memory storing programming instructions for execution by the at least one processor, the programming instructions, when executed by the at least one processor, causing the computing system to perform the following operations: (Kurian teaches the system may be implemented on computer systems including processors and memory storing computer executable instructions. Kurian at ¶¶ [0038] – [0039].).
Regarding claim 11, Kurian-Khoury teaches all the limitations of claim 10 as laid out above. Kurian-Khoury, however, may not explicitly disclose all of the limitations of claim 11.
Huffman in the same field of endeavor teaches wherein the plurality of voice characteristics include at least one of: speaker age; speaker language; and speaker gender. ([0084], gender, ethnicity, age, etc.). Therefore, it would have been obvious at the time the application was filed to use the above feature of Huffman with the system of Kurian in view of Khoury, in order to provide generative system that combine both data diversity and privacy protection.
As per claims 15-18, Kurian teaches a computer program product residing on a non-transitory computer readable medium having programming instructions stored thereon which, when executed by at least one processor of a system, cause the system to perform the following operations: (Kurian teaches the system may be implemented on computer systems including processors and memory storing computer executable instructions (Kurian at ¶¶ [0038] – [0039]). The remaining steps are rejected under the same rationale as applied to the method steps of rejected claims 8-11.
Regarding claim 22, Kurian in view of Khoury may not explicitly disclose wherein each of the synthetic biometric profiles indicates statistical distributions of biometric characteristics across two or more of the natural biometric profiles, collectively. Huffman in the same field of endeavor teaches a voice-to-voice conversion system that uses its synthetic biometric profiles to generate voices of a large cohort of speakers that share particular traits such as ethnicity, accent, age, gender, etc. ([0082]- [0087]. Accordingly, the synthetic biometric profiles indicate statistical distributions of biometric characteristics across two or more of the natural biometric profiles, collectively. Therefore, it would have been obvious at the time the application was filed to use the above feature of Huffman with the system of Kurian in view of Khoury, in order to increase the efficiency and accuracy of the selection of results.
Regarding claim 26, Kurian in view of Khoury may not explicitly disclose wherein the cohort of biometric profiles, defined by the synthetic biometric profiles, reflect distributions of voice characteristics amongst a subset of the known speakers within an age range ([0084]). Huffman in the same field of endeavor teaches a voice-to-voice conversion system, wherein the cohort of biometric profiles, defined by the synthetic biometric profiles, reflect distributions of voice characteristics amongst a subset of the known speakers within an age range ([0084], wherein a characteristic line distinguishes between cohorts of voices based on gender, ethnicity, age, etc.). Therefore, it would have been obvious at the time the application was filed to use the above feature of Huffman with the system of Kurian in view of Khoury, in order to provide generative system that combine both data diversity and privacy protection.
Regarding claim 27, Kurian in view of Khoury may not explicitly disclose wherein the cohort of biometric profiles, defined by the synthetic biometric profiles, reflect distributions of voice characteristics amongst a subset of the known speakers that speak a common language. Huffman in the same field of endeavor teaches a voice-to-voice conversion system, wherein the cohort of biometric profiles, defined by the synthetic biometric profiles, reflect distributions of voice characteristics amongst a subset of the known speakers within an age range ([0084], wherein a characteristic line distinguishes between cohorts of voices based on gender, ethnicity, age, etc. and also [0059], wherein speakers are speaking English as a common language but different accents). Therefore, it would have been obvious at the time the application was filed to use the above feature of Huffman with the system of Kurian in view of Khoury, in order to provide generative system that combine both data diversity and privacy protection.
Regarding claim 26, Kurian in view of Khoury may not explicitly disclose wherein the cohort of biometric profiles, defined by the synthetic biometric profiles, reflect distributions of voice characteristics amongst a subset of the known speakers having a common gender. Huffman in the same field of endeavor teaches a voice-to-voice conversion system, wherein the cohort of biometric profiles, defined by the synthetic biometric profiles, reflect distributions of voice characteristics amongst a subset of the known speakers having a common gender ([0084], wherein a characteristic line distinguishes between cohorts of voices based on gender). Therefore, it would have been obvious at the time the application was filed to use the above feature of Huffman with the system of Kurian in view of Khoury, in order to provide generative system that combine both data diversity and privacy protection.
Conclusion
5. The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. See PTO-892.
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 ABDELALI SERROU whose telephone number is (571)272-7638. The examiner can normally be reached M-F 9 Am - 5 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, Pierre-Louis Desir can be reached at 571-272-7799. 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.
/ABDELALI SERROU/ Primary Examiner, Art Unit 2659
Prosecution Insights