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 .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 10/10/2024, 11/07/2024 07/07/2025, 07/09/2025, 07/30/2025, 08/14/2025 ,08/26/2025, 11/17/2025, 12/04/2025 and 12/19/2025 are considered by the examiner.
Drawings
The information disclosure statement (IDS) submitted on 11/09/2023 is considered by the examiner.
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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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.
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.
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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claim(s) 1-2, 5, 8-12, 15, and 18-20, are rejected under 35 U.S.C. 103 as being unpatentable over Hennig et al. (US 20220328050 A1) in view of Caldwell (US 2019/0311722 A1).
Regarding claim 1, Hennig et al. teach: A computer-implemented method for authenticating users based on speech of audio signals, comprising : obtaining, by a computer, responsive to successfully authenticating a user according to a first authentication of a user (designated voice print 130 associated with the user account 104) using authentication credentials (authentication credentials), an audio speech signal of the user for a second authentication (voice verification of current voice print) ([0021] As some examples, the authenticator component can utilize authentication credentials in the form of usernames, passwords, passcodes (e.g., sending a passcode via message to a phone number or email account associated with the user account), or personal identification numbers (PINs), personal questions relating to the user, other biometric information (e.g., fingerprint information, eye or iris biometric information, or facial biometric information), or other desired authentication and verification techniques to attempt to verify and authenticate the unidentified user as the user associated with the account. [0025] In accordance with various embodiments, in connection with performing authentication processes and managing access to the user accounts 104, information 106, transactions 108, applications 110, and/or services 112, the authenticator component 102 can employ voice authentication and verification, wherein a user can speak (e.g., to a service representative of the service entity or to the system 100), and the authenticator component 102 can analyze the voice of the user, and based at least in part on the voice biometrics of the voice, the authenticator component 102 can verify whether the voice is actually the voice of the user, as opposed to a fraudulent attempt by another user (e.g., adversarial or malicious user) to represent as being the user. [0026] In some embodiments, in connection with a user account 104 associated with a user, the authenticator component 102 can receive voice data representative of a voice purporting to be the user from an unidentified user 114, for example, via a communication device 116 (e.g., via a communication network with which the authenticator component 102 and communication device 116 can be associated.); extracting, by the computer, a plurality of features (set of characteristics (e.g., voice patterns, digital signal zero-crossings rate associated with the voice, and/or other characteristics) associated with the received voice data) from the audio speech signal; generating, by the computer, a plurality of initial liveness scores (voice resemblance scores) using the plurality of features (generate voice resemblance scores (e.g., similarity scores) based at least in part on a comparison of a set of characteristics associated with received voice data) by applying a machine-learning architecture to the plurality of features extracted from the audio speech signal, the machine-learning architecture trained to generate each initial liveness score (voice resemblance score) using a corresponding subset of features of the plurality of features extracted from the audio speech signal ([0032] The voice verification component 118 can comprise a fraud detector component 136 that can analyze (e.g., compare) the set of characteristics (e.g., attributes) of the voice data (and/or the current voice print associated with the set of characteristics) and the set of characteristics associated with the designated voice print (e.g., 130) and the set of previous voice prints (e.g., 124) associated with the user account 104 of the user. [0034] To facilitate authentication and verification, the voice verification component 118 can comprise a voice resemblance scoring component 202 that can determine (e.g., calculate) and generate voice resemblance scores (e.g., similarity scores) based at least in part on a comparison of a set of characteristics associated with received voice data and a designated voice print (e.g., designated voice print 130) and/or a comparison of the set of characteristics associated with the received voice data and a set of previous voice prints (e.g., set of previous voice prints 124). [0035] For instance, with regard to the voice data received from the unidentified user 114, the voice resemblance scoring component 202 can compare the set of characteristics (e.g., voice patterns, digital signal zero-crossings rate associated with the voice, and/or other characteristics) associated with the received voice data to the set of characteristics associated with the designated voice print 130 associated with the user account 104 of the user. Based at least in part on the results of such comparison, the voice resemblance scoring component 202 can determine a first voice resemblance score.);generating, by the computer, based upon the plurality of initial liveness scores generated using the plurality of features extracted from the audio speech signal, a liveness score (voice resemblance score is lower than or equal to (e.g., does not satisfy) the second threshold voice resemblance score) indicating a likelihood that the user is a human (actual and live voice of the user) speaker; and executing execute, by the computer, using the liveness score, the second authentication to determine whether to permit the user access based upon comparing the liveness score against a threshold ([0038] In response to the determination that the voice data is a match to the designated voice print 130, the voice verification component 118 can perform the second level of authentication and verification. The voice verification component 118 can access the set of previous voice prints 124 associated with the user account 104 of the user from the voice print repository 122. The voice resemblance scoring component 202 can compare the set of characteristics associated with the received voice data to the respective sets of characteristics associated with respective previous voice prints of the set of previous voice prints 124. Based at least in part on the results of such comparison, the voice resemblance scoring component 202 can determine a second voice resemblance score, which can correspond to a level of similarity between the set of characteristics associated with the received voice data and set of characteristics associated with a previous voice print of the set of previous voice prints 124. [0040] If, based at least in part on the comparison results, the fraud detector component 136 determines that the second voice resemblance score is lower than or equal to (e.g., does not satisfy) the second threshold voice resemblance score, the fraud detector component 136 can determine that the received voice data is not too close of a match to the set of previous voice prints 124, and, accordingly, can determine that the voice associated with the received voice data is valid or verified as being the voice of the user. In response to determining that the voice has been verified as being the voice of the user (e.g., actual and live voice of the user), the authenticator component 102 can authenticate the user (e.g., can recognize the previously unidentified user 114 as the user) and/or associated communication device 116, and can grant the user and/or communication device 116 access to the user account 104 and/or associated information 106, transactions 108, applications 110, and/or services 112.).
Hennig et al. do not teach: extracting, by the computer, a plurality of features from the audio speech signal, including one or more audio background features; at least one initial liveness score generated using at least one subset of features having the one or more audio background features.
Caldwell teaches: extracting, by the computer, a plurality of features from the audio speech signal, including one or more audio background features (identify the acoustic code received in the audio input signal (i.e. from the acoustic code components)); at least one initial liveness score generated using at least one subset of features having the one or more audio background features(acoustic code generator 144 outputs an acoustic code associated with the user) ([0029] In operation of the voice interaction device 110, the user 102 launches the voice authentication client 140 on the user device 130 and is presented with a pass phrase to repeat. The user 102 repeats the pass phrase which is received by the microphone 112. [0030] When the user is prompted to repeat the pass phrase, the acoustic code generator 144 outputs an acoustic code associated with the user (e.g., acoustic code 168 stored in authentication database 166) through the speaker 136, which is also picked up by the microphone 112. The audio input processing components 114 separate the audio input signal into speech and acoustic code components for authentication by voice authentication components 118. The voice processing 120 and voice authentication components 118 identify the pass phrase from the speech component, and compare voice features from the speech component to the stored user voice feature to determine a confidence score. In various embodiments, the user voice features may be stored locally by the voice interaction device 110 and/or other devices. [0031] The voice authentication components 118 also identify the acoustic code received in the audio input signal (i.e. from the acoustic code components) and match the received acoustic code against the user's unique code (e.g., such as stored acoustic codes 168). Authentication will be found if both the user's voice and the acoustic code are authenticated by the voice authentication components 118. In one embodiment, authentication is found if the acoustic code matches the user's acoustic code and a degree of confidence of a voice match between the received speech and user voice features exceeds a threshold selected for a given implementation.)
Therefore, it would have been obvious to one of ordinary skilled in the art at the time of the invention was made for Hennig et al. to include the teaching of Caldwell above, identify a pass phrase and an acoustic code from an input audio signal, in order to authenticate a user using both speech and acoustic code by voice authentication components.
Regarding claims 2 and 12, Hennig et al. teach: The method of claim 1, wherein obtaining the audio speech signal includes presenting, by the computer, a prompt to direct the user to provide the audio speech signal, responsive to authenticating the user in the first authentication using the authentication credentials (See rejection of claim 1, and [0029] The authenticator component 102 can generate a previous voice print based at least in part on a phrase spoken by the user when attempting to authenticate with the authenticator component 102 or other words spoken by the user during an interaction, wherein the phrase can be something the authenticator component 102 or a service representative may request the user to speak as part of the authentication and verification process.).
Regarding claims 5 and 15: The method of claim 1, wherein the plurality of initial liveness scores comprises at least one of: (i) a first score identifying background change, (ii) a second score identifying passive liveness of the speech signal of the speaker, or (iii) a third score identifying repetition of speech within the speech signal (See rejection of claim 1).
Regarding claims 8 and 18, Hennig et al. teach: The method of claim 1, wherein performing the second authentication includes performing, by the computer, the second authentication to restrict the user access to a resource, responsive to the liveness score not satisfying the threshold (See rejection of claim 1 and 0044] In some embodiments, if the fraud detector component 136 determines that the received voice data is too close of a match to a previous voice print of the set of previous voice prints 124, the fraud detector component 136 can determine that the received voice data is fraudulent (without performing a third level of authentication and verification), when in accordance with the defined authentication criteria. For example, if the second voice resemblance score corresponds to a 100% or almost 100% (e.g., 99%) probability, which can indicate that there can be a high probability that the received voice data is fraudulent, the fraud detector component 136 can determine that it is not desirable (e.g., not necessary or useful) to utilize further resources (e.g., computing resources, time resources, or other resources) to perform the third level of authentication and verification to attempt to verify the received voice data and associated unidentified user 114. Accordingly, the authenticator component 102 can determine that the unidentified user 114 and/or associated communication device 116 are not to be authenticated with regard to the user account 104, and the unidentified user 114 and/or associated communication device 116 can be denied access to the user account 104 and/or associated information 106, transactions 108, applications 110, and/or services 112.).
Regarding claims 9 and 19, Hennig et al. teach: The method of claim 1, wherein performing the second authentication includes performing, by the computer, the second authentication to permit the user access to a resource, responsive to the liveness score satisfying the threshold (See rejection of claim 1 and [0040] If, based at least in part on the comparison results, the fraud detector component 136 determines that the second voice resemblance score is lower than or equal to (e.g., does not satisfy) the second threshold voice resemblance score, the fraud detector component 136 can determine that the received voice data is not too close of a match to the set of previous voice prints 124, and, accordingly, can determine that the voice associated with the received voice data is valid or verified as being the voice of the user. In response to determining that the voice has been verified as being the voice of the user (e.g., actual and live voice of the user), the authenticator component 102 can authenticate the user (e.g., can recognize the previously unidentified user 114 as the user) and/or associated communication device 116, and can grant the user and/or communication device 116 access to the user account 104 and/or associated information 106, transactions 108, applications 110, and/or services 112.).
Regarding claims 10 and 20, Hennig et al. teach: The method of claim 1, further comprising generating, by the computer, an indication of a result of the second authentication indicating whether to permit the user access based on the liveness score (See rejection of claim 1 and [0040] If, based at least in part on the comparison results, the fraud detector component 136 determines that the second voice resemblance score is lower than or equal to (e.g., does not satisfy) the second threshold voice resemblance score, the fraud detector component 136 can determine that the received voice data is not too close of a match to the set of previous voice prints 124, and, accordingly, can determine that the voice associated with the received voice data is valid or verified as being the voice of the user. In response to determining that the voice has been verified as being the voice of the user (e.g., actual and live voice of the user), the authenticator component 102 can authenticate the user (e.g., can recognize the previously unidentified user 114 as the user) and/or associated communication device 116, and can grant the user and/or communication device 116 access to the user account 104 and/or associated information 106, transactions 108, applications 110, and/or services 112.).
Regarding claim 11, Hennig et al. teach: A system for authenticating callers using speech of audio signals in calls, comprising: a computer comprising one or more processors configured to ([0010] The voice interaction device further comprises a voice processor operable to detect speech and execute associated voice commands, including user voice authentication.): obtain responsive to successfully authenticating a user according to a first authentication using authentication credentials, an audio speech signal of the user for a second authentication; extract a plurality of features from the audio speech signal, including one or more audio background features; generate a plurality of initial liveness scores using the plurality of features by applying a machine-learning architecture to the plurality of features, the machine-learning architecture trained to generate each initial liveness score using a corresponding subset of features of the plurality features extracted from the audio speech signal, including at least one initial liveness score generated using at least one subset of features having the one or more audio background features from the plurality of features; generate based upon the plurality of initial liveness scores generated using the plurality of features extracted from the audio speech signal, a liveness score indicating a likelihood that the user is a human speaker; and execute the second authentication to determine whether to permit the user access based upon comparing the liveness score against a threshold (See rejection of claim 1).
Claim(s) 3, 6-7, and 13, 16-17, are rejected under 35 U.S.C. 103 as being unpatentable over Hennig et al. in view of Caldwell further in view of Gross (US 2009/0319274 A1).
Regarding claims 3 and 13, Hennig et al. in view of Caldwell teach: The method of claim 1 wherein generating the liveness score includes applying, by the computer, a machine-learning architecture to the plurality of features to generate the liveness score(See rejection of claim 1)
Hennig et al. in view of Caldwell, however does not teach: wherein the machine-learning architecture is trained on a plurality of examples, each example identifying a second plurality of features and a label indicating one of human speech or machine-generated speech.
Gross teaches: wherein the machine-learning architecture is trained on a plurality of examples, each example identifying a second plurality of features and a label indicating one of human speech or machine-generated speech ([0010]Thus, some speaker-verification technology has ways of testing for "liveness." They specifically analyze for acoustic patterns suggesting that the voice has been recorded using a process called anti-spoofing. [0016] A first aspect of the invention concerns a method of identifying a source of data input to a computing system comprising: receiving speech utterance from an entity related to randomly selected challenge text; wherein the challenge text represents a selected set of one more contiguous words which when articulated have a measurable difference in acoustical characteristics between a reference human voice and a reference computer synthesized voice that exceeds a target threshold; processing the speech utterance with the computing system to compute first acoustical characteristics of the entity; and generating a determination of whether the speech utterance originated from a machine or a human. [0017] Also preferred embodiments may have the steps: soliciting utterances from a plurality of separate computing machines to determine their respective acoustical characteristics; and storing the plurality of associated acoustical characteristics in a database of known computing entities. Multiple samples of individual challenge sentences are preferably collected. The challenge text is preferably selected in part based on a difference in time for rendering such text into audible form by a human and a computing machine.).
Therefore, it would have been obvious to one of ordinary skilled in the art at the time of the invention was made for Hennig et al. in view of Caldwell to include the teaching of Gross above automatically determining a difference in between a reference human voice and a reference computer synthesized voice acoustical characteristics, in order to determine whether the speech utterance originated from a machine or a human.
Regarding claims 6 and 16: The method of claim 1 wherein extracting the plurality of features includes applying, by the computer, the machine-learning architecture to generate the plurality of features including a set of embeddings (feature characteristics with the user's voice OR both visual and audible CAPTCHAs used in a hybrid challenge) representing spoofing artifacts in the audio speech signal (See rejection of claim 1 and Gross teaching: [0010] Thus, some speaker-verification technology has ways of testing for "liveness." They specifically analyze for acoustic patterns suggesting that the voice has been recorded using a process called anti-spoofing. [0011] To date, therefore, while verification systems have been used for distinguishing between humans, they have been designed or employed on a limited basis for the purpose of distinguishing between a computer speaking and a human speaking as part of a CAPTCHA type tester/analyzer. [0107] For example, both visual and audible CAPTCHAs could be used in a hybrid challenge system. An entity could be presented simultaneously with a number of visual distinct word CAPTCHAs arranged in the form of a sentence. The entity is then required to read an entire sentence of words that are each visually confounded, thus increasing the chances that a computing system will fail to process such data in a reasonable time frame.).
Regarding claims 7 and 17: The method of claim 6, wherein generating the liveness score includes applying, by the computer, the machine-learning architecture to the set of embedding representing the spoofing artifacts to determine the liveness score (See rejection of claim 6).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. The prior art of records Lesso (US 2019/0114497 A1) teach: DETECTION OF LIVENESS .
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MOHAMMAD K ISLAM whose telephone number is (571)270-5878. The examiner can normally be reached Monday -Friday, EST (IFP).
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, Paras Shah can be reached at 571-270-1650. 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.
/MOHAMMAD K ISLAM/Primary Examiner, Art Unit 2653