DETAILED ACTION
Notice of Pre-AIA or AIA Status
1. 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/Amendments
2. With respect to Rejection under 35 U.S.C. § 101, Applicant argues on the Remark that
“In Ex Parte Desjardins, USPTO Director Squires endorsed the use of the Alice Mayo Test in determining the subject matter eligibility of patent applications under 35 U.S.C. § 101. However, Director Squires highlighted that under Step 2A, Prong Two of that test, it is important to look to the Federal Circuit's precedential decision in Enfish, LLC v. Microsoft Corp., 822 F.3d 1327 (Fed. Cir. 2016) which held that “[s]oftware can make non-abstract improvements to computer technology, just as hardware can.” Ex Parte Desjardins, Appeal 2024-000567 at 8 (citing to Enfish at 1336).
In the opinion, Director Squires cited to Enfish in evaluating whether the claimed AI invention at issue in Ex Parte Desjardins was an improvement in computer technology. He concluded that it was. The opinion asserted that the modifications to the underlying AI training algorithm in that application “allow[] artificial intelligence (AI) systems to 'us[e] less of their storage capacity’ and enable[] ‘reduced system complexity.’” Id. at 9 (Citing internally to the Desjardin application specification). Squires further opined that “this case demonstrates that § 102, 103 and 112 are the traditional and appropriate tools to limit patent protection to its proper scope. These statutory provisions should be the focus of examination.” Id. at 10.
Applying that same analysis to the present application demonstrates that the claimed
invention provides a non-abstract improvement to computer technology that should be allowed under Section 101. In particular, claim 1 of the application makes clear that the invention calculates an air conduction sound signal, a bone conduction sound signal, an air conduction feature quantity, and a bone conduction feature quantity, and combines the air and bone conduction feature quantities into a single target feature quantity. While the air conduction feature quantity is an N-dimensional vector and a bone conduction feature quantity is an M-dimensional vector, the combined target feature quantity is a vector with dimension less than N+M. See Instant Application at [0058] (“As described above, since the air conduction feature quantity is an N-dimensional vector and the bone conduction feature quantity is an M-dimensional vector”); see also Instant Application at [0059] (“The combination feature quantity, however, may be a vector of less than N+M-dimensions. That is, the number of dimensions of the combination feature quantity may be less than N+M.”). As in Ex Parte Desjardins, this is an improvement to the underlying AI modeling, because it combines multiple feature vectors into a single, smaller input feature vector for the AI model, enabling the model to use less storage capacity while reducing overall system complexity. Thus, the invention is subject matter eligible under Step 2A, Prong Two of the Alice Mayo test.”
In response, Examiner respectfully notes that Desjardins involves a particular approach to training a machine learning model leading to an improvement in the field of artificial intelligent model training. The present process differs in that the machine leaning model is not improved as a tool, but uses the machine learning model as to a tool to implement an otherwise abstract idea. Thus, the claimed invention does not improve the use of the machine learning model as a tool and differs from Desjardins. See MPEP 2106.05(f) for discussion of mere instructions to implement an otherwise abstract idea.
Step 2A, Prong Two: MPEP 2106.05(f) provides the following considerations for determining whether a claim simply recites a judicial exception with the words “apply it” (or an equivalent), such as mere instructions to implement an abstract idea on a computer: (1) whether the claim recites only the idea of a solution or outcome i.e., the claim fails to recite details of how a solution to a problem is accomplished; (2) whether the claim involves computers or other machinery merely as a tool to perform an existing process; and (3) the particularity or generality of the application of the judicial exception. The machine learning model is used without placing any limits on how the machine learning model functions. Rather, the limitation only recites the outcome of “authenticate the target person”. The recitation of “using a machine learning model” in the limitation also merely indicates a field of use or technological environment in which the judicial exception is performed. Although the additional element “using a machine learning model” limits the judicial exception, this type of limitation merely confines the use of the abstract idea to a particular technological environment (machine learning) and thus fails to add an inventive concepts to the claims. See MPEP 2106.05(h).
The present process takes the target feature quantity (i.e., combination of the air conduction quantity and the bone conduction feature quantity) having a less than
N+M-dimensional vector as input to the machine leaning model to authenticate the target person. The present process does not involve in training the machine learning model in any aspects to reduce the number of dimensions of the combination feature quantity.
Even when view in combination, these additional elements do not integrate the recited judicial exception into a practical application (Step 2A, Prong Two: NO), and the claim is directed to the judicial exception. (Step 2A: YES).
Applicant’s arguments are not persuasive, and thus for these reasons, Examiner respectfully disagrees. Consequently, the 101 abstract idea rejection is maintained.
Claim Rejections - 35 USC § 101
3. 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.
4. Claims 1-6 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
Claim 1 recites
“1. (Currently Amended) An authentication apparatus comprising:
at least one memory configured to store instructions; and
at least one processor configured to execute the instructions to:
calculate, from an air conduction sound signal indicating an air conduction sound of a voice of a target person and a bone conduction sound signal indicating a bone conduction sound of the voice of the target person, an air conduction feature quantity that is a feature quantity of the air conduction sound signal and is an N-dimensional vector and a bone conduction feature quantity that is a feature quantity of the bone conduction sound signal and is an M-dimensional vector, and that calculates a target feature quantity that is a less than N+M-dimensional vector and is a feature quantity of the voice of the target person by combining the air conduction feature quantity and the bone conduction feature quantity; and
authenticate using a machine learning model the target person on the basis of the target feature quantity.”
Claims 1 and 6 recite substantially the same concept but do so in the context of an apparatus and a method.
The limitations recited in the independent claims as drafted covers a mathematical concept. More specifically, it relates to calculating an air conduction feature quantity, a bone conduction feature quantity, combining the air conduction feature quantity and the bone conduction quantity and authenticating a user based on the combination of the air conduction feature quantity and the bone conduction quantity. Claims 1 and 6 recites the number of dimensions of the air conduction feature quantity, the bone conduction feature quantity and the combination feature quantity.
Claim 4 recites
“4. (Currently Amended) An authentication apparatus comprising:
at least one memory configured to store instructions; and
at least one processor configured to execute instructions to:
calculate, from an air conduction sound signal indicating an air conduction sound of a voice of a target person and a bone conduction sound signal indicating a bone conduction sound of the voice of the target person, an air conduction feature quantity that is a feature quantity of the air conduction sound signal and is an N-dimensional vector and a difference feature quantity that is a feature quantity of a difference between a frequency spectrum of the air conduction sound signal and a frequency spectrum of the bone conduction sound signal; and
authenticate using a machine learning model the target person on the basis of the air conduction feature quantity and the difference feature quantity.”
The limitations recited in the independent claims as drafted covers a mathematical concept. More specifically, it relates to calculating an air conduction feature quantify and a bone conduction feature quantity, a different feature quantity between the air conduction feature quantity and the bone conduction feature quantity and authenticating a user based on the basis of the air conduction feature quantity and the different quantity. Claim 4 recites the number of dimensions of the air conduction feature quantity.
The judicial exception is not integrated into a practical application. In particular, claim recites additional element of “at least one memory”, “at least one process” and “using a machine learning model”. The additional element(s) or combination of elements such as a memory and/or a processor in the claim(s) other than the abstract idea per se amount(s) to no more than (i) mere instructions to implement the idea on a computer, and/or (ii) recitation of generic computer structure that serves to perform generic computer functions that are well-understood, routine, and conventional activities previously known to the pertinent industry. Viewed as a whole, these additional claim element(s) do not provide meaningful limitation(s) to transform the abstract idea into a patent eligible application of the abstract idea such that the claim(s) amounts to significantly more than the abstract idea itself. Therefore, the claim(s) are rejected under 35 U.S.C. 101 as being directed to non-statutory subject matter. There is further no improvement to the computing device other than calculating the different spectrum and authenticating the user based on the different spectrum. The mere recitation of a memory, a processor and/or the like is akin of adding the word “apply it” and/or “use it” with a computer in conjunction with the abstract idea. The paragraph [0030] discloses “The arithmetic apparatus 31 includes at least one of a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), and a FPGA (Field Programmable Gate Array), for example. The arithmetic apparatus 31 reads a computer program. For example, the arithmetic apparatus 31 may read a computer program stored in the storage apparatus 32. For example, the arithmetic apparatus 31 may read a computer program stored by a computer-readable and non-transitory recording medium, by using a not-illustrated recording medium reading apparatus provided in the authentication apparatus 3 (e.g., the input apparatus 34 described later). The arithmetic apparatus 31 may acquire (i.e., download or read) a computer program from a not-illustrate apparatus disposed outside the authentication apparatus 3, through the communication apparatus 33 (or another communication apparatus). The arithmetic apparatus 31 executes the read computer program. Consequently, a logical functional block for performing an operation to be performed by the authentication apparatus 3 (e.g., the authentication operation described above) is realized or implemented in the arithmetic apparatus 31. That is, the arithmetic apparatus 31 is allowed to function as a controller for realizing or implementing the logical functional block for performing an operation (in other words, a process) to be performed by the authentication apparatus 3.)
As filed in the specification, the computer is listed as a general-purpose computer and are mainly used as an application thereof. Accordingly, these additional elements do not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea.
The machine learning model is used without placing any limits on how the machine learning model functions. Rather, the limitation only recites the outcome of “authenticate the target person”. The recitation of “using a machine learning model” in the limitation also merely indicates a field of use or technological environment in which the judicial exception is performed. Although the additional element “using a machine learning model” limits the judicial exception, this type of limitation merely confines the use of the abstract idea to a particular technological environment (machine learning) and thus fails to add an inventive concepts to the claims. See MPEP 2106.05(h).
The present process takes the target feature quantity (i.e., combination of the air conduction quantity and the bone conduction feature quantity) having a less than
N+M-dimensional vector as input to the machine leaning model to authenticate the target person. The present process does not involve in training the machine learning model in any aspects to reduce the number of dimensions of the combination feature quantity.
Even when view in combination, these additional elements do not integrate the recited judicial exception into a practical application (Step 2A, Prong Two: NO), and the claim is directed to the judicial exception. (Step 2A: YES).
The dependent claims further do not remedy the issues noted above. More specifically, Claim 2 recites using a neural network in calculating the target feature quantity. The neural network is recited at a high level of generality; the limitation only recites the outcome of “the target feature quantity” without placing any limits on how the neural network functions. Claim 3 recites further limitation of the difference feature quantity and the number of dimensions of the difference feature quantity. Claim 3 recites the machine learning model to authenticate the target person. Please, see the analysis related to the machine learning model and the number of dimensions of the vector in Claim 1. Claim 5 recites the authentication includes two processes. The first process is based on the air conduction feature quantity and the second process is based on the different feature quantity. Claims 2-3 and 5 recites a processor. The mere recitation of a processor and/or the like is akin of adding the word “apply it” and/or “use it” with a computer in conjunction with the abstract idea.
For at least the supra provided reasons, claims 1-6 are rejected under 35 U.S.C. 101 as being directed to non-statutory subject matter.
Allowable Subject Matter
5. Claims 1-6 are allowed in view of the prior art of record. The claims stand rejected under 101 Abstract idea, and for the application to pass to allowance this rejection need to be overcome. Any amendments to overcome the 101 rejection that results in any change in scope require further search and/or consideration in order to determine it allowability.
The following is a statement of reasons for the indication of allowable subject matter: the prior art(s) taken alone or in combination fail(s) to teach the following element(s) in combination with the other recited elements in the claim(s).
“calculate, from an air conduction sound signal indicating an air conduction sound of a voice of a target person and a bone conduction sound signal indicating a bone conduction sound of the voice of the target person, an air conduction feature quantity that is a feature quantity of the air conduction sound signal and is an N-dimensional vector and a bone conduction feature quantity that is a feature quantity of the bone conduction sound signal and is an M-dimensional vector, and that calculates a target feature quantity that is a less than N+M-dimensional vector and is a feature quantity of the voice of the target person by combining the air conduction feature quantity and the bone conduction feature quantity; and
authenticate using a machine learning model the target person on the basis of the target feature quantity.” as recited in Claim 1.
Claim 6 recites the similar features as Claim 1.
“calculate, from an air conduction sound signal indicating an air conduction sound of a voice of a target person and a bone conduction sound signal indicating a bone conduction sound of the voice of the target person, an air conduction feature quantity that is a feature quantity of the air conduction sound signal and is an N-dimensional vector and a difference feature quantity that is a feature quantity of a difference between a frequency spectrum of the air conduction sound signal and a frequency spectrum of the bone conduction sound signal; and
authenticate using a machine learning model the target person on the basis of the air conduction feature quantity and the difference feature quantity.” as recited in Claim 4.
The closest prior arts found as following.
a. Kameyama et al. (US 2006/0046684 A1.) Kameyama et al. disclose a method and a system for authenticating a user based on a differential spectrum of the air conduction and the bone conduction sound (Kameyama et al. Fig. 23A frequency spectrum of an air conduction sound, Fig. 23B frequency spectrum of a bone conduction sound, Fig. 23 C the differential spectrum of the air conduction sound and the bone conduction sound, [0142] As another alternative authentication, the user terminal 1 may read the resultant differential spectrum of the air conduction sound and the bone conduction sound, as shown in FIG. 23C, after the voice recognition. Then the user terminal 1 compares the resultant differential spectrum with the standard differential spectrum 223 shown in FIG. 17. If the user terminal 1 determines that both of the spectra are almost identical to each other, then the user terminal 1 sets the authentication flag as "allowed". If the difference between the spectra is not within a permissible range, then the user terminal 1 sets the authentication flag as “not allowed”.) Kameyama et al. authenticates the target user by calculating the frequency spectrum of the air conduction sound, the frequency spectrum of the bone conduction sound, and differential spectrum between the frequency spectrum of the air conduction sound and the frequency spectrum of the bone conduction sound. Kameyama et al. compares the calculated differential spectrum with the standard different spectrum to authenticate the target user. Kameyama et al. disclose authenticating the target user on the basis of the claimed target feature quantity (i.e., combining the air conduction feature quantity and the bone conduction feature quantity.) However, Kameyama et al. does not use a machine learning model in authenticating the target user. Kameyama et al. does not disclose the air conduction feature quantify is an N-dimensional vector, the bone conduction feature quantity is M-dimensional vector, and the targe feature quantify is a less than N+M-dimensional vector as recited in Claims 1 and 6. Kameyama et al. does not disclose the air conduction feature quantify is an N-dimensional vector and using the a machine learning model to authenticate on the basis to the frequency spectrum of the air conduction sound and the calculated different spectrum as recited in Claim 4. Thus, Kameyama et al. fail to teach and/or suggest the allowable subject matter.
b. Zhang et al. (US 2021/0256979 A1.) In this reference, Zhang et al. disclose collecting a first voice component by an air conduction microphone, collecting a second voice component by a bone conduction microphone to authenticate a target person (Zhang et al. [0025] With reference to any one of the second aspect and the possible design methods of the second aspect, in a fifth possible design method of the second aspect, the performing, by the wearable device, voiceprint recognition on the first voice component and the second voice component includes: determining, by the wearable device, whether the first voice component matches a first voiceprint model of an authorized user, where the first voiceprint model is used to reflect an audio feature that is of the authorized user and that is collected by the first voice sensor; and determining, by the wearable device, whether the second voice component matches a second voiceprint model of the authorized user, where the second voiceprint model is used to reflect an audio feature that is of the authorized user and that is collected by the second voice sensor; and [0026] after the performing, by the wearable device, voiceprint recognition on the first voice component and the second voice component, the method further includes: if the first voice component matches the first voiceprint model of the authorized user, and the second voice component matches the second voiceprint model of the authorized user, determining, by the wearable device, that a voicing user is an authorized user, or otherwise, determining, by the wearable device, that the voicing user is an unauthorized user.) Zhang et al. uses both the first voice component corrected by the air conduction microphone and the second voice component correct by the bone conduction microphone. Zhang et al. teach the voicing user is the authorized user when the first voice component matches the first voiceprint model and the second voice component matches the second voiceprint model. Zhang et al. teach using a machine learning algorithm in establishing a background model of voiceprint recognition. However, Zhang et al. does not combine the first voice component and the second voice component in order to calculate a voice component and using a machine learning model to authenticate the user in the basis of the voice component. Zhang et al. does not determine the difference between the first voice component and the second voice component. Thus, Zhang et al. fail to teach and/or suggest the allowable subject matter.
c. Lesso (US 2019/0012448 A1.) In this reference, Lesso disclose using an air-conducted voice signal and a bone-conducted audio signal to authenticate a target person (Lesso Fig. 3a, [0079] Thus FIG. 3a shows an embodiment in which the bone-conducted voice biometric algorithm is combined with an air-conducted voice biometric algorithm, [0080] FIG. 3b shows a system 350 in which the bone-conducted voice biometric algorithm is combined with an ear biometric algorithm, [0088] The extracted feature(s) are passed to a biometric module 366, which performs a biometric process on them, in a similar manner to the biometric module 318 described above. For example, the biometric module 366 may perform a biometric enrolment, in which the extracted features (or parameters derived therefrom) are stored as part of biometric data which is characteristic of the individual. The biometric data may be stored within the system or remote from the system (and accessible securely by the biometric module 366). Such stored data may be known as an “ear print”. In another example, the biometric module 366 may perform a biometric authentication, and compare the one or more extracted features to corresponding features in the stored ear print (or multiple stored ear prints, [0093] Again, the authentication may be carried out on the combination of data in multiple different ways. In one embodiment separate authentication algorithms may be carried out on each of the sets of data, and separate authentication scores acquired from each of the sets of data. The scores may then be combined to generate an overall score, indicating the overall likelihood that the user is an authorized user, with the authentication decision being taken on this score (e.g. by comparing the score to a threshold). In an alternative embodiment, the individual biometric scores may be handled separately (e.g., compared to separate thresholds) and individual authentication decisions being taken on each score. Overall authentication is then based on a combination of the decisions. For example, failure at any one of the authentication algorithms may result in failure of the authentication overall. Thus, if the bone-conducted authentication algorithm results in a positive authentication, but the ear authentication algorithm results in a rejection, the user may be rejected overall.) Lesso combines the air-conducted voice biometric algorithm with the bone-conducted voice biometric algorithm to authenticate a target person. Lesso combines scores in both algorithm to generate an overall score indicating the overall likelihood that the target user is an authorized user. Lesso also discloses the individual biometric scores are handled separately (e.g., compared to separate thresholds) and individual authentication decisions taken on each score. Overall authentication is then based on a combination of the decisions. For example, failure at any one of the authentication algorithms may result in failure of the authentication overall. Lesso uses quality measure of the air-conducted data and the quality measure of the bone-conducted data. However, Lesso does not teach combines the quality measure of the air-conducted data and the quality measure of the bone-conducted data to generate the overall quality measure, wherein the quality measure of the air-conducted data is an N-dimensional vector, the quality measure of the bone-conducted data is an M-dimensional vector and the overall quality measure is a less than N+M-dimensional vector as recited in Claims 1 and 6. Lesso does not disclose calculating the difference between the quality measure of the air-conducted data and the quality measure of the bone-conducted data and authenticate the target user on the basis of the quality measure of the air-conducted data the calculated difference as recited in Claim 4. Thus, Lesso fail to teach and/or suggest the allowable subject matter.
Conclusion
6. The prior art made of record and not relied upon is considered pertinent to application’s disclosure. See PTO-892.
a. Li et al. (US 2024/0203408 A1.) In this reference, Li et al. utilizes the input bone conduction registration signal and the input air conduction registration signal in verifying a user.
b. Klemme et al. (US 2019/0045298 A1.) In this reference, Klemme et al. utilizes the bone conduction analyzer to perform a voice authentication analysis of the sound data to verify the user and to permit or deny access to the user application.
c. Lesso (US 2019/0012447 A1.) In this reference, Lesso discloses a method for biometric processes. The biometric processes in Lesso includes authentication of the air conducted voice signal and the bone conducted voice signal.
7. THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a). 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.
8. Any inquiry concerning this communication or earlier communications from the examiner should be directed to THUYKHANH LE whose telephone number is (571)272-6429. The examiner can normally be reached Mon-Fri: 9am-5pm.
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, Andrew C. Flanders can be reached on 571-272-7516. 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.
/THUYKHANH LE/Primary Examiner, Art Unit 2655