DETAILED ACTION
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
This is in response to Applicant’s remarks and request for reconsideration filed 8/31/2026.
Claim Interpretation
The claims frequently use the alternative language as shown below. Applicant is reminded that a prior art reference need only disclose ONE of the listed alternatives to meet that limitation. The claim is satisfied (and thus anticipated/rendered obvious) if any single option in the list is disclosed, because the applicant has claimed them disjunctively.
Claim 1: “one or more tasks”
Claim 2: “at least one of [CPU, GPU, …]”
Claim 3: “one of [SIP, PSTN, POTS, …]”
Claim 4: “at least one of [telemarketing, robocall, fraud]”
Claim 6: “at least one of [first/second/third portion]”
Claim 9: “at least one of [word count, phoneme count, inter-word timing, pitch, background noise]”
Claim 12: “one or more of the at least one of [metrics]”
Claim 13: “at least one of [metrics]” (grouping)
Claim 14: “at least one of [entities, groups, characteristics]”
Claim 15: “at least one of [block, busy, message, update, train]”
Claim Objections
Claim 16 is objected to because of the following informalities:
Claim 16 recites “obtained data associated with a plurality of calls,” which appears to omit the preposition “from” in line 9. The recitation is understood to be intended to read “obtained from data associated with a plurality of calls.”
Appropriate correction is required.
Claim Rejections - 35 USC § 112(a)
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claims 1 – 19 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the enablement requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to enable one skilled in the art to which it pertains, or with which it is most nearly connected, to make and/or use the invention.
The claims require that “first abstracted raw data” be obtained from a portion of call “data associated with the first call” (claims 1, 7, 16, 17), wherein that portion is “of a first predetermined duration” (claim 6) and wherein the abstracted raw data comprises “word count data, phoneme count data, inter-word timing data, voice pitch estimation data, or background noise data” (claim 9). Obtaining such data necessarily requires access to the spoken audio content transmitted by the originating party. The specification confirms this, disclosing that the analyzed portion is an extract of spoken speech — for example, “This is Molly from xCloud Support. We have found some suspicious activity in your xCloud account” (specification ¶[0075]) - and that the portion is of a duration on the order of 5 to 120 seconds or longer (specification ¶¶[0016], [0053]).
However, the specification repeatedly and consistently locates this spoken audio content within a call request, rather than within an established, answered media session. Is the call answered or not answered? Specifically:
“a call request 205 for a call from a first originating party to a first destination party may include call data including a message 210a” (specification ¶[0063]);
“extract portion 315 from a first message 310, which may be contained within a call request for a call from a first originating party to a first destination party” (specification ¶[0075]);
“the call requests may include SIP data indicating a request to initiate a SIP-based media communication session between a calling party … and a called party” (specification ¶[0048]).
The recited tasks reinforce that the claimed analysis and determination are completed prior to the call being answered. Claims 15 and 19 recite, as a task performed after the determination that the call is a suspected unsolicited or unwanted communication, “responding to a call request that is associated with the first call with a busy signal” (see also specification ¶¶[0037], [0041]). Returning a busy signal in response to a call request is, by its nature, an action taken during call setup and before the call is answered or connected. Likewise, “blocking the first call” (claims 15, 19) is described as an action taken to prevent the call from proceeding.
The specification therefore requires that a substantial quantity of the originating party’s spoken audio be captured, abstracted, compared against a corpus, and adjudicated - all before the call has been answered and before the audio would conventionally be transmitted. The specification does not disclose any mechanism by which this is accomplished. The specification nowhere describes:
Whether the first call is answered by any entity or computing system prior to capture of the call data. The specification does not use the term “answer,” “answered,” “connected,” or any equivalent, and does not identify any point in the call lifecycle at which the recited capture occurs.
How spoken audio content of 5 to 120 seconds or longer is obtained from a “call request” or from signaling data. The specification identifies the signaling protocols employed as SIP, and identifies the call request as “SIP data indicating a request to initiate a SIP-based media communication session” (specification ¶[0048]). The specification does not disclose how the originating party’s spoken message is conveyed within, or extracted from, such a request, nor does it identify any protocol field, header, body, or extension by which spoken audio of the recited duration would accompany a request to initiate a session.
Any alternative capture architecture - such as answering the call, auto-answering the call, bridging a media agent onto an established session, playing a greeting to elicit the message, recording the call to a voicemail or media server, or otherwise establishing a media path - by which the required audio content could be obtained consistent with the claimed pre-answer disposition of the call.
While the specification states generally that detection “may be determined in near-real-time” (specification ¶[0082]), and identifies the computing system as possibly including “a voice activity detection (‘VAD’) system,” a “correlation engine,” a “call server,” or a “media gateway controller” (specification ¶[0050]), these disclosures identify only where processing may occur. They do not disclose how the spoken audio content is acquired relative to answering the call, which is the critical unenabled feature.
The above demonstrates that the specification does not enable the full scope of the claimed invention without undue experimentation.
The rejection applies to all claims, as each requires obtaining and abstracting call data from the first call. The deficiency is most acute in claims 15 and 19, which expressly recite performing the task of “responding to a call request that is associated with the first call with a busy signal” after the determination has been made from the abstracted call audio data. The specification does not enable a method in which the originating party’s spoken audio is captured and analyzed and a busy signal is thereafter returned in response to the associated call request.
Claims 6, 7, 8, and 9 are further affected, as each presupposes access to audio content of substantial duration and substantive acoustic and structural characteristics (predetermined-duration portions; abstraction of raw data; word count, phoneme count, inter-word timing, voice pitch estimation, and background noise data), the acquisition of which is not enabled for the reasons stated above.
Claim Rejections - 35 USC § 112(b)
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 6 – 15 and 19 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 6 recites “the portion of the call data” and “the call data.” Claim 1 recites neither call data nor any portion thereof, reciting only “data associated with the first call.” There is Insufficient antecedent basis.
Claim 7 recites “the portion of the call data,” which lacks antecedent basis for the reason stated above, and “the first originating party,” where claim 1 recites only “an originating party.” It is unclear whether the latter refers to that party, to one of the recited plurality of originating parties, or to a further party.
Claim 8 recites “the portion of the call data,” “the first originating party,” and “the call data,” each lacking antecedent basis for the reasons stated above.
Claim 9 recites “the at least one of word count data, phoneme count data, inter-word timing data, voice pitch estimation data, or background noise data,” twice. Claim 1 does not recite that either the first abstracted raw data or the plurality of abstracted raw data comprises any of these data types. Insufficient antecedent basis.
Claim 10 is indefinite by virtue of its dependency from claim 9.
Claim 11 recites “the plurality of second scores.” No plurality of second scores is previously recited in claim 11, claim 9, or claim 1. Insufficient antecedent basis. Claim 11 is further indefinite by virtue of its dependency from claim 9.
Claim 12 recites “the plurality of second scores” and “the at least one of word count data, phoneme count data, inter-word timing data, voice pitch estimation data, or background noise data,” each lacking antecedent basis for the reasons stated above.
Claim 13 recites the same two limitations lacking antecedent basis. Claim 13 is further indefinite in reciting groups “characterized by similarities in each of the at least one of” the recited data types. The phrase “each of” requires a plurality, whereas “at least one of” requires only one. It cannot be determined whether similarity in all, one, or some subset of the recited data types is required.
Claim 14 is indefinite by virtue of its dependency from claim 13.
Claim 15 recites “updating a corpus of suspected robocalls to include the first abstracted raw data to the plurality of abstracted raw data.” The construction “include X to Y” is grammatically indeterminate, and the intended operation cannot be ascertained. Claim 15 further recites “training rapid fraud determination,” which does not identify what is trained, by what method, or what renders a determination “rapid.” Note that “rapid” is a relative term.
Claim 15 recites, a task performed after the determination:
“responding to a call request that is associated with the first call with a busy signal”. The determination, however, is made from abstracted data derived from the call’s audio content. This creates a logical conflict:
If the determination requires audio, then the call must have been answered/connected to obtain the audio. A busy signal cannot meaningfully be returned to the associated call request.
Additionally, the relationship between “the first call” (claim 1) and “a call request that is associated with the first call” (claims 15) is unclear. It cannot be ascertained whether the call request is the same transaction as the first call.
Claim 19 is indefinite for the reasons stated with respect to claim 15.
Double Patenting
The nonstatutory double patenting rejection is based on a judicially created doctrine grounded in public policy (a policy reflected in the statute) so as to prevent the unjustified or improper timewise extension of the “right to exclude” granted by a patent and to prevent possible harassment by multiple assignees. A nonstatutory double patenting rejection is appropriate where the conflicting claims are not identical, but at least one examined application claim is not patentably distinct from the reference claim(s) because the examined application claim is either anticipated by, or would have been obvious over, the reference claim(s). See, e.g., In re Berg, 140 F.3d 1428, 46 USPQ2d 1226 (Fed. Cir. 1998); In re Goodman, 11 F.3d 1046, 29 USPQ2d 2010 (Fed. Cir. 1993); In re Longi, 759 F.2d 887, 225 USPQ 645 (Fed. Cir. 1985); In re Van Ornum, 686 F.2d 937, 214 USPQ 761 (CCPA 1982); In re Vogel, 422 F.2d 438, 164 USPQ 619 (CCPA 1970); In re Thorington, 418 F.2d 528, 163 USPQ 644 (CCPA 1969).
A timely filed terminal disclaimer in compliance with 37 CFR 1.321(c) or 1.321(d) may be used to overcome an actual or provisional rejection based on nonstatutory double patenting provided the reference application or patent either is shown to be commonly owned with the examined application, or claims an invention made as a result of activities undertaken within the scope of a joint research agreement. See MPEP § 717.02 for applications subject to examination under the first inventor to file provisions of the AIA as explained in MPEP § 2159. See MPEP § 2146 et seq. for applications not subject to examination under the first inventor to file provisions of the AIA . A terminal disclaimer must be signed in compliance with 37 CFR 1.321(b).
The filing of a terminal disclaimer by itself is not a complete reply to a nonstatutory double patenting (NSDP) rejection. A complete reply requires that the terminal disclaimer be accompanied by a reply requesting reconsideration of the prior Office action. Even where the NSDP rejection is provisional the reply must be complete. See MPEP § 804, subsection I.B.1. For a reply to a non-final Office action, see 37 CFR 1.111(a). For a reply to final Office action, see 37 CFR 1.113(c). A request for reconsideration while not provided for in 37 CFR 1.113(c) may be filed after final for consideration. See MPEP §§ 706.07(e) and 714.13.
The USPTO Internet website contains terminal disclaimer forms which may be used. Please visit www.uspto.gov/patent/patents-forms. The actual filing date of the application in which the form is filed determines what form (e.g., PTO/SB/25, PTO/SB/26, PTO/AIA /25, or PTO/AIA /26) should be used. A web-based eTerminal Disclaimer may be filled out completely online using web-screens. An eTerminal Disclaimer that meets all requirements is auto-processed and approved immediately upon submission. For more information about eTerminal Disclaimers, refer to www.uspto.gov/patents/apply/applying-online/eterminal-disclaimer.
Claims 1 – 19 are rejected on the ground of nonstatutory double patenting as being unpatentable over claims 1- 19 of prior U.S. Patent No. 12177382. Although the claims at issue are not identical, they are not patentably distinct from each other because:
The difference between the independent claims 1, 16 and 17 in the current application and the corresponding independent claims 1, 16 and 17 in the Patent is the deletion of the following limitation from the application claims:
based on a determination that the first abstracted raw data is not indicative of the first call being a suspected unsolicited or unwanted communication, connecting the first call to 12 the first destination party
The current claims are broader and thus there are rejected under obviousness-type double patenting over the previously patented narrower claims. In re Van Ornum, 686 F.2d 937, 214 USPQ 761. The dependent claims correspond to those in the Patent.
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 1–5, 9 and 15–19 are rejected under 35 U.S.C. § 101 because the claimed invention is directed to a judicial exception (an abstract idea) without significantly more.
Step 1, the claims are directed to a process (method) , a machine (apparatus) and a machine (system). Thus, the claims fall within the statutory categories of 35 U.S.C. § 101.
Step 2A, Prong One - the claims recite an abstract idea
Claim 1 recites:
“comparing … first abstracted raw data, obtained from data associated with the first call, with each of a plurality of abstracted raw data, obtained from data associated with a plurality of calls …”
“determining … whether the first abstracted raw data is indicative of the first call being a suspected unsolicited or unwanted communication, based at least in part on the comparison …”
These limitations recite a mental process — a category of abstract idea identified in the 2019 Revised Patent Subject Matter Eligibility Guidance (MPEP 2106.04(a)(2)(III)). Specifically:
The “comparing” step recites an act of observation and comparison of information.
The “determining” step recites an act of evaluation and judgment — concluding, from the comparison, that a call is suspected to be unwanted.
Comparing two sets of information and forming a judgment based on that comparison are functions that can be characterized as practicable in the human mind or with the aid of pen and paper. Claim 1 recites these steps at a functional, result-oriented level of generality: the claim does not specify how the comparison is performed, nor does it specify the particular content or structure of the “abstracted raw data” being compared, nor any particular technique by which the determination is reached. Note that the determination may, for example, be based non background noise and a human may be able to make that determination.
Claim 1, as drafted, therefore encompasses the bare concept of comparing information derived from a call against information derived from other calls and judging the call to be unwanted.
Claims 16 and 17 recite the same “compare” and “determine” functions as claim 1, expressed as instructions executed by a processor. Reciting an abstract idea as computer-implemented instructions does not remove the recitation of the judicial exception (MPEP 2106.04(a)). Claims 16 and 17 likewise recite a mental process.
Dependent Claims 2–5, 9, 15, 18, and 19 depend from, and therefore incorporate, the abstract idea recited in claims 1, 16, or 17. They are addressed under Prong Two.
Step 2A, Prong Two – No Integration Into a Practical Application
The claims are evaluated to determine whether the additional elements, beyond the judicial exception, integrate the exception into a practical application.
In Independent Claim 1, the additional limitations are:
The limitation “a computing system” (performing the comparing and determining) recited a high level of generality; amounts to mere instructions to implement the abstract idea on a generic computer (MPEP 2106.05(f)).
The limitation “receiving … a first call from an originating party intended for a first destination party” recites mere data gathering / receipt of the information to be analyzed (MPEP 2106.05(g)).
The limitation “performing … one or more tasks” recites without any specificity as to what the tasks are. This very generic, result-oriented recitation constitutes insignificant post-solution activity and does not impose a meaningful limit on the exception (MPEP 2106.05(g)).
Considered individually and as an ordered combination, these additional elements do not: (i) improve the functioning of a computer or another technology or technical field (MPEP 2106.05(a)); (ii) apply the exception with or by use of a particular machine (MPEP 2106.05(b)); (iii) effect a transformation of a particular article (MPEP 2106.05©); or (iv) apply the exception in any other meaningful way beyond generally linking it to a technological environment (MPEP 2106.05(e), (h)).
Rather, the additional elements do no more than instruct a generic computing system to carry out the abstract comparison and judgment, with generic data receipt on the front end (such as a human listening to a background noise) and an undefined action on the back end.
Prong Two: claim 1 does not integrate the abstract idea into a practical application.
Claims 16 and 17 add only “at least one processor” and a “non-transitory computer readable medium … having stored thereon computer software comprising a set of instructions.” These are generic computer components recited at a high level of generality and represent the paradigmatic case of “apply it” on a computer (MPEP 2106.05(f)). Reciting a generic processor and memory storing instructions does not integrate the exception into a practical application.
Claims 2 and 18 recite that the computing system comprises “at least one of one or more central processing units (‘CPUs’), one or more graphics processing units (‘GPUs’), … a call server, … a centralized server, or a distributed computing system.” Each recited alternative is a generic, conventional computing component or arrangement. Reciting a list of generic hardware options, in the alternative, does not supply a particular machine or a technical improvement; it merely specifies generic components on which the abstract idea is to be applied (MPEP 2106.05(b), (f)).
Claim 3 recites that the first call comprises “one of a session initiation protocol (‘SIP’) -based communication, a public switched telephone network (‘PSTN’) communication, or a plain old telephone service (‘POTS’) communication,” including VoIP, IP-based video, or instant message over IP. This limitation merely generally links the abstract idea to a particular technological environment or field of use - telephony networks - which does not integrate the exception into a practical application (MPEP 2106.05(h)).
Claim 4 recites that the suspected unwanted communication comprises “at least one of a telemarketing communication, a robocall communication, or a communication instigating fraudulent activity.” Claim 5 recites that the communication “comprises a pre-recorded message.”
These limitations characterize the subject matter being evaluated — that is, they narrow the field of application of the abstract judgment — without reciting any technical mechanism by which the evaluation is performed. Narrowing the abstract idea to a particular type of target communication does not render the claim eligible (MPEP 2106.05(h)).
For claim 9, which recites many alternatives, the human can recognize the background noise. Similar to claim 1, it does not integrate the exception into a practical application
Claims 15 and 19 recite that the “one or more tasks” comprise “at least one of: blocking the first call; responding to a call request … with a busy signal; sending a message to the first destination party …; updating a corpus of suspected robocalls …; or training rapid fraud determination.”
Each recited alternative constitutes insignificant extra-solution or post-solution activity - an action taken after the abstract determination has been made (MPEP 2106.05(g)). Blocking a call, returning a busy signal, sending a notification, or adding an entry to a list are conventional, routine telecommunications and data-storage operations. Such post-solution activity, recited generically and in the alternative, does not impose a meaningful limitation on the abstract idea or reflect an improvement to the underlying technology.
Step 2B — the Claims do not Provide an Inventive Concept
The additional elements identified above are reevaluated to determine whether they amount to significantly more than the judicial exception.
The recited computing components — a “computing system,” “at least one processor,” a “non-transitory computer readable medium,” and the generic hardware alternatives of claims 2 and 18 — perform only their routine, conventional, and well-understood functions: receiving data, storing data, comparing data, and outputting a result. As the specification itself acknowledges, the invention may be implemented using conventional general-purpose computing hardware, including “one or more general-purpose processors,” conventional storage devices, and standard communications subsystems (spec ¶¶[0097]–[0099]), and may be practiced on a “general purpose personal computer,” servers, or workstations “running any of a variety of commercially-available UNIX™ or UNIX-like operating systems” (spec ¶[0108]).
Reliance on generic computer components to perform generic computer functions does not supply an inventive concept. Alice Corp. v. CLS Bank Int’l, 573 U.S. 208, 223–26 (2014); MPEP 2106.05(f).
Likewise, the data-receipt step and the generically recited post-determination “tasks” (claims 15, 19) represent conventional activity in the telecommunications field and do not, individually or in combination, transform the claims into a patent-eligible application of the abstract idea (MPEP 2106.05(d), (g)).
Considered as an ordered combination, the elements of claims 1–5, 9 and 15–19 add nothing beyond what would be expected when implementing the abstract comparison-and-judgment on a generic computer within a conventional telephony environment.
Claim Rejections - 35 USC § 102
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 the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claims 1–5, 7–10, and 15–19 are rejected under 35 U.S.C. § 102(a)(1) and/or § 102(a)(2) as being anticipated by Li (US 2021/0136200 A1).
Regarding claim 1, Li discloses A method, comprising:
Receiving a call - “receiving, by a computing system, a first call from an originating party intended for a first destination party.” Li discloses receiving audio input representing telephone calls placed to a call recipient. Li teaches that “[t]he system receives live or recorded audio containing real or simulated human speech” (¶[0010]) and that the system “receives audio (e.g., a recorded or live phone call that has not previously been analyzed by the system)” (¶[0014]). Li further discloses that the voice biometrics detection system 115 “receive[s] information associated with calls made by one or more callers 110 … to the advertisers 112” (¶[0026]), where a caller (originating party) places a call directed to a recipient (first destination party). (See Li, ¶¶[0010], [0014], [0026]; FIGS. 1, 3B, block 355.)
Comparing abstracted raw data - “comparing, using the computing system, first abstracted raw data, obtained from data associated with the first call, with each of a plurality of abstracted raw data, obtained from data associated with a plurality of calls from a corresponding plurality of originating parties.” Li discloses generating a voice fingerprint (first abstracted raw data) from the audio of the received call and comparing it against a plurality of stored voice fingerprints (plurality of abstracted raw data) corresponding to a plurality of callers. Li teaches that “the system extracts unique characteristics for each speaker that can be used to generate voice fingerprints” (¶[0008]) and that “the system searches for voice fingerprints in the known speaker dataset that match the generated voice fingerprint … the system calculates a probability that the generated voice fingerprint matches one or more voice fingerprints stored in the known speaker dataset” (¶[0014]). Li further discloses that “[t]he system can compare a generated voice fingerprint to stored datasets of known callers and caller type types (e.g., robocallers, spam callers, legitimate callers, etc.)” (¶[0007]). The voice fingerprint is expressed as “compressed and/or uncompressed data vectors or arrays of one or more voice biometrics” (¶[0037]), constituting data abstracted from the raw audio.
The stored dataset comprises fingerprints from a “plurality of known callers” (¶[0008]), satisfying “a plurality of abstracted raw data … from a plurality of calls from a corresponding plurality of originating parties.” (See Li, ¶¶[0007]–[0009], [0014], [0037]; FIG. 3B, blocks 360, 365.)
Determining suspected unwanted communication - “determining, using the computing system, whether the first abstracted raw data is indicative of the first call being a suspected unsolicited or unwanted communication, based at least in part on the comparison.” Li discloses determining, based on the fingerprint comparison, whether the call is a robocall or spam call. Li teaches that “[u]pon determining a match between the generated voice fingerprint from the call audio data and one or more voice fingerprints stored in the … dataset, the system determines that the speaker in the call audio data and the identified speaker in the dataset … are the same,” and uses the associated classification (e.g., “a robocaller, a spam caller”) to characterize the call (¶[0014]). Li further discloses that “[i]f the similarity score exceeds a threshold score, then the system determines that there is a match” (¶[0015]), and characterizes the call as being associated with a robocaller (Claim 1: “characterizing a voice fingerprint within the dataset as being associated with a robocaller”). (See Li, ¶¶[0014]–[0015]; FIG. 3B, blocks 365, 370, 375.)
Performing one or more tasks - “based on a determination that the first abstracted raw data is indicative of the first call being a suspected unsolicited or unwanted communication, performing, using the computing system, one or more tasks.” Li discloses taking corrective action upon determining the call is a robocall/spam call. Li teaches “the system may take various actions … to request confirmation from a call recipient … or take other corrective action” including that “the system may terminate the call, display a warning, request a call recipient to confirm that the call is spam” (¶[0007]). Li further discloses “taking corrective steps such as ‘allowlisting’ or ‘denylisting’ phone numbers … [or] handling the call differently” (¶[0013]). (See Li, ¶¶[0007], [0013]; FIG. 3B, block 380.)
Claim 2 recites “…wherein the computing system comprises at least one of ……”. This limitation is recited in the alternative (“at least one of”). Li discloses the computing system comprising “one or more central processing units (CPU) 200 for executing software” (¶[0032]; FIG. 2). Li additionally discloses a “server, or other computing system” and a “distributed computing environment” (¶¶[0024]–[0025]).
Claim 3 recites “…wherein the first call comprises one of …..”. This limitation is recited in the alternative (“one of”). Li discloses that calls traverse networks including “a Voice over Internet Protocol (VoIP) network, a cellular telecommunications network, a public-switched telephone network (PSTN)” (¶[0027]).
Claim 4 recites “…wherein the suspected unsolicited or unwanted communication comprises at least one of ….”. This limitation is recited in the alternative. Li is directed to detecting “robocalls and other spam calls” (¶[0002]) and expressly addresses calls “used for fraudulent purposes” (¶[0002]).
Claim 5 recites “… comprises a pre-recorded message.” Li discloses that “[r]obocalls are typically prerecorded so that they can be played repeatedly” (¶[0002]) and addresses machine-generated calls that “mimic or simulate the human voice” (¶[0008]).
Claim 7 recites “…further comprising: abstracting, using the computing system, raw data from the portion of the call data associated with the first call from the first originating party to produce the first abstracted raw data.” Li discloses extracting voice biometrics from the received audio to generate the voice fingerprint: “the system extracts voice biometrics from the received audio using AI models and data processing techniques” (¶[0010]) and “extract[s] one or more voice biometrics for speakers in the audio” (¶[0035]; FIG. 3B, block 360).
Claim 8 recites “… without transcription of communication contained within the call data and without use of speech-to-text or text-to-speech functionality …”. Li discloses that its voice-biometric analysis is content-independent and does not rely on transcription or speech-to-text. Li teaches that “the system can identify a speaker even though the spoken words or sentences may differ from speech used to generate a voice fingerprint because voice biometrics are largely consistent with respect to a speaker. In other words, the system can extract and use biometrics to generate voice fingerprints that identify the same speaker regardless of the content of the received speech” (¶[0009]). The fingerprint is derived from acoustic biometrics (volume, pitch, speaking rate, pauses) rather than transcribed text (¶[0008]).
Claim 9 recites “…comparing … comprises determining … one or more similarities … based at least in part on comparison of the at least one of word count data, phoneme count data, inter-word timing data, voice pitch estimation data, or background noise data …” This limitation is recited in the alternative (“at least one of”). Li discloses that the voice biometrics used to generate and compare fingerprints include “pitch” (¶¶[0008], [0035]), which corresponds to the claimed “voice pitch estimation data.” Li further discloses “pauses between each utterance” (¶[0008]), relevant to “inter-word timing data.” Thus, Li discloses at least one recited alternative (voice pitch estimation data), and the comparison of fingerprints is based on these biometrics via a similarity score (¶[0015]),
Claim 10 recites “…generating … a similarity score based on the determined one or more similarities … wherein determining whether the first abstracted raw data is indicative … comprises determining that the generated similarity score is greater than a predetermined matching threshold value.” Li discloses generating a similarity score and comparing it to a threshold: “the system generates a similarity score. If the similarity score exceeds a threshold score, then the system determines that there is a match” (¶[0015]). Li further discloses that “[t]hresholds for matching can be configurable or based on empirical data” (¶[0009]) and that a match is found when the probability is “greater than 85%-90%” (¶[0009]).
Claim 15 recites “…wherein the one or more tasks comprise at least one of: blocking the first call; responding to a call request … with a busy signal; sending a message to the first destination party indicating that the first call is a suspected unsolicited or unwanted communication; updating a corpus of suspected robocalls to include the first abstracted raw data …; or training rapid fraud determination.”
This limitation is recited in the alternative. Li discloses “blocking the first call” - the system “may automatically block or flag denylisted callers” and “automatically disconnect the phone call” (¶¶[0016], [0019], [0047]). Li also discloses “sending a message to the first destination party” - transmitting a warning to the call recipient (¶[0007], ¶[0047]). Li further discloses “updating a corpus of suspected robocalls” — “add[ing] the speaker voice fingerprint associated with such calls” to the dataset and creating “a new known caller entry in the maintained dataset” (¶¶[0017], [0051]).
Li therefore discloses multiple recited alternatives (blocking the call, sending a message to the recipient, and updating the corpus of fingerprints).
Independent claim 16 recites an apparatus. Li discloses the voice biometrics detection system 115 comprising “one or more central processing units (CPU) 200 for executing software stored in the storage area 230” (¶[0032]). Li further discloses that “[a]spects of the system described herein may be stored or distributed on tangible, non-transitory computer-readable media” (¶[0025]) and that “computer-executable instructions … may be stored in memory, such as random access memory (RAM), read-only memory (ROM)” (¶[0024]). The storage area 230 stores executable “software modules and data that, when executed or operated on by a processor, perform certain of the methods or functions described herein” (¶[0030]). (See Li, ¶¶[0024]–[0025], [0030], [0032]; FIG. 2.) Claim 16 is rejected for the same reasons as claim 1.
Independent Claim 17 recites a system. This is disclosed by Li for the same reasons set forth above with respect to claim 16. Li discloses the voice biometrics detection system 115 as a “system” (¶[0030]) comprising CPU 200 (¶[0032]) and non-transitory computer-readable media storing executable instructions (¶¶[0024]–[0025], [0030]). (See Li, ¶¶[0024]–[0025], [0030], [0032]; FIG. 2.) The claim is also rejected for the same reasons as claim 1.
Claim 18 is rejected for the same reasons as Claim 2 above.
Claim 19 is rejected for the same reasons as Claim 2 above.
Claims 1 - 5, 7 - 8, 10, and 15 – 19 are rejected under 35 U.S.C. § 102(a)(1) and/or § 102(a)(2) as being anticipated by Gallagher (US 2021/0092223 A1).
Claim 1 recites receiving a call. Gallagher discloses “[w]hen an incoming phone call is received at a receiving party’s device within a communication network” (¶[0016]), where the call originates from an incoming caller (originating party) directed to a receiving party (first destination party). ¶¶[0016], [0034].)
For the claimed comparing abstracted raw data, Gallagher discloses creating an acoustic fingerprint (first abstracted raw data) from the incoming caller’s audio stream and comparing it against a plurality of stored acoustic fingerprints (plurality of abstracted raw data): “an acoustic fingerprint may be created and compared with acoustic fingerprints stored within the acoustic profile data store” (¶[0029]). The acoustic fingerprint comprises abstracted data - “a number of hashes created during a specific call. Hashes are any two frequency points and the distance in time between them” (¶[0037]). The AP data store 118 stores “previously created acoustic fingerprints of phone calls” from a plurality of calls and callers (¶[0035]), satisfying “each of a plurality of abstracted raw data … from a plurality of calls from a corresponding plurality of originating parties.” See ¶¶[0015]–[0016], [0029], [0035], [0037].
For the claimed determining suspected unwanted communication, Gallagher discloses that “[a]t the point a match is found, then it is determined that the call source is a machine, e.g., a robocall” (¶[0016]). The matching of the created acoustic fingerprint to a stored fingerprint constitutes determining that the first abstracted raw data is indicative of a suspected unwanted communication (robocall). See ¶¶[0015]–[0016], [0029].
For the claimed performing one or more tasks, Gallagher discloses that upon detection, “[t]he communication network can apply one or more call handling procedures such as, for example, notifying a traceback group …, ending the phone call, and/or providing a warning message to the call recipient” (¶[0016]). See ¶¶[0016]–[0017].
For Claim 2 and 18, (alternative hardware list): Gallagher discloses CPU/processor and FPGA/ASIC (¶[0045])
For Claim 3 (alternative SIP/PSTN/POTS), Gallagher discloses VoIP and PSTN networks (¶[0033])
Claim 4 recites “…at least one of a telemarketing communication, a robocall communication, or a communication instigating fraudulent activity.” Recited in the alternative. Gallagher is directed to detecting “robocalls” (Abstract; ¶[0002]) and addresses “potentially fraudulent phone calls” (Abstract).
Claim 5 recites “…comprises a pre-recorded message.” Gallagher discloses detecting calls where “the originating party is a pre-recorded message and/or a machine” and the robocaller “begins playing its own pre-recorded message” (¶¶[0028]–[0029]).
Claim 7 recites “…abstracting … raw data from the portion of the call data … to produce the first abstracted raw data.” Gallagher discloses the acoustic fingerprinter creating an acoustic fingerprint from the caller’s audio stream: “[t]he acoustic fingerprinter 116 screens the audio stream from the originating party … and creates an acoustic fingerprint, in real-time based on that audio stream” (¶[0035]).
Claim 8 recites “…without transcription … and without use of speech-to-text or text-to-speech functionality …” Gallagher discloses that acoustic fingerprinting is content-independent: “Acoustic fingerprinting is language independent and an audio recording cannot be reverse engineered from an acoustic fingerprint” (¶[0029]; see also ¶[0015]). The fingerprint is derived from frequency-point hashes (¶[0037]), not from transcription or speech-to-text.
For Claim 10 (similarity score > threshold): Gallagher discloses matching against a threshold via robocall scores (¶[0025]).
Claim 15 recites “…at least one of: blocking the first call; … sending a message to the first destination party …; updating a corpus of suspected robocalls …; or training rapid fraud determination.”
Recited in the alternative. Gallagher discloses “ending the phone call” (blocking) and “providing a warning message to the call recipient” (sending a message to the first destination party) (¶[0016]). Gallagher further discloses “updating a corpus of suspected robocalls” — storing the fingerprint in the AP data store and the number in the robocall number data store (¶¶[0016], [0036]). Gallagher discloses multiple recited alternatives.
Claim 16 is rejected for the same reasons as claim 1. Claim 16 recites a processor and non-transitory medium. Gallagher discloses “one or more processors 204” and “one or more non-transitory storage media comprising instructions stored thereon, the instructions being executable by the one or more processors” (¶[0045]; Claim 18). (See Gallagher, ¶¶[0044]–[0048]; FIG. 2.)
Claim 17 is rejected for the same reasons as claim 1. Claim 17 recites a first processor and first non-transitory medium. Gallagher discloses the robocall detection service 110 / acoustic profiler platform 112 implemented on computing device 202 comprising processor(s) 204 and non-transitory memory 208 storing executable instructions (¶¶[0044]–[0049]; FIG. 2).
Claim 19 recites “…at least one of: blocking the first call; … sending a message to the first destination party …; updating a corpus of suspected robocalls …; or training rapid fraud determination.”
Recited in the alternative; disclosed by Gallagher for the same reasons set forth above with respect to claim 15.
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.
Claims 11, 12, and 13 are rejected under 35 U.S.C. § 103 as being unpatentable over Li (US 2021/0136200 A1) in view of Hwang et al. (US 8,141,152 B1, “Hwang”).
Claim 11 recites “…determining the one or more similarities … comprises: generating … a first score based on the first abstracted raw data; comparing … the first score with each of the plurality of second scores each based on a corresponding one of the plurality of abstracted raw data; and generating … a similarity score for each of the plurality of second scores based on the comparison …”
Li discloses generating a value from the first abstracted raw data (voice fingerprint) and comparing it against stored values to generate similarity scores. Li teaches “calculating a probability that the generated voice fingerprint matches one or more voice fingerprints stored in the known speaker dataset” and computes “a probability of match for each speaker voice fingerprint” (¶¶[0014], [0043]).
To the extent Li does not expressly disclose the recited score-to-score architecture (a first score compared against a plurality of second scores), Hwang discloses this.
Hwang teaches “calculating a first distance from a first of the plurality of feature vectors to each feature vector in an associated spam feature database” (Claim 1) and computing a comparison result for each stored item. Hwang expressly discloses generating quantified comparison values (distances) between an incoming abstracted representation and each of a plurality of stored abstracted representations (col. 13, lines 50–65; FIG. 6, block 624, and Claim 1)
It would have been obvious to one of ordinary skill in the art at the time of the invention to implement Li’s fingerprint comparison using the score-based distance comparison of Hwang because both references address detecting unsolicited communications by comparing an abstracted representation of an incoming call against a plurality of stored abstracted representations of known spam/robocalls. Hwang provides an efficient, quantified framework for such comparison (col. 13), and applying it to Li’s fingerprint dataset would predictably yield an accurate match determination with a reasonable expectation of success. (See Li, ¶¶[0014], [0043]; Hwang, Claim 1, col. 13.)
Claim 12 recites “…prior to comparing the first score with each of the plurality of second scores, pruning … the plurality of abstracted raw data to reduce a number of second scores … that require comparison, by filtering out abstracted raw data … based at least in part on a difference in one or more of the [metrics].”
Li’s comparison does not expressly disclose pruning.
Hwang discloses reducing the number of comparisons prior to full comparison: “if it is too computationally costly to compare a particular message with all known spam messages one-by-one, as is usually the case in a very large spam database, then feature vectors of known spam messages may be ‘clustered’ first based on their proximity to one another in the feature vector space, then the incoming message may be compared with the cluster heads” (col. 14, lines 5-18). Hwang thereby reduces the number of required comparisons based on differences/proximity in the abstracted representations.
It would have been obvious to prune Li’s fingerprint dataset as taught by Hwang to reduce computational cost. Hwang expressly identifies the motivation: avoiding the computational expense of “compar[ing] a particular message with all known spam messages one-by-one” in “a very large spam database” (col. 14). Li likewise contemplates “large datasets” (¶[0021]), so one of ordinary skill would have been motivated to apply Hwang’s pruning to improve scalability with a reasonable expectation of success. (See Hwang, col. 14; Li, ¶[0021].)
Claim 13 recites “…grouping … the plurality of abstracted raw data into one or more groups characterized by similarities in each of the [metrics]; wherein pruning … comprises filtering out abstracted raw data … based at least in part on the one or more groups.”
Hwang discloses grouping by similarity: “feature vectors of known spam messages may be ‘clustered’ first based on their proximity to one another in the feature vector space” (col. 14). The clusters constitute “one or more groups characterized by similarities,” and the subsequent comparison to cluster heads constitutes filtering based on the groups.
Same as claim 12. Grouping/clustering to reduce comparisons is expressly taught by Hwang for scalability, and applying it to Li’s dataset would predictably improve efficiency. (See Hwang, col. 14.)
Claims 6 is rejected under 35 U.S.C. § 103 as being unpatentable over Li (US 2021/0136200 A1) in view of Gallagher et al. (US 2021/0092223 A1, “Gallagher”).
Claim 6 recites “…wherein the portion of the call data of the first call comprises at least one of: a first portion of the call data, the first portion … being of a first predetermined duration; a second portion … overlapping with the first portion, without an initial portion that is of a second predetermined duration; or a third portion … after any of the first portion or the second portion, the third portion being of a third predetermined duration.”
This limitation is recited in the alternative (“at least one of”); disclosure of a single alternative suffices. Li discloses analyzing a portion of the call audio, teaching that “the system is able to identify sufficient biometrics from several seconds (e.g., 3-5 or more seconds) of received audio to characterize speakers” (¶[0038]). Li thus discloses using a first portion of the call data of a predetermined duration.
To the extent Li does not expressly recite a portion of a defined predetermined duration extracted for analysis, Gallagher discloses creating an acoustic fingerprint from “at least a portion of the audio stream created by the incoming caller” (¶[0014]) and processing defined segments of the audio.
It would have been obvious to analyze a first portion of predetermined duration, as suggested by both Li (¶[0038], “several seconds”) and Gallagher (¶[0014], “at least a portion”), because analyzing a bounded portion of call audio conserves processing resources and enables near-real-time detection, an express objective of both references (Li ¶[0042]; Gallagher ¶[0016]). One of ordinary skill would have combined these teachings with a reasonable expectation of success. (See Li, ¶¶[0038], [0042]; Gallagher, ¶¶[0014], [0016].)
Claim 14 is rejected under 35 U.S.C. § 103 as being unpatentable over Li (US 2021/0136200 A1) in view of Hwang et al. (US 8,141,152 B1, “Hwang”), as applied to claim 13 above, and further in view of Gallagher et al. (US 2021/0092223 A1, “Gallagher”).
The combination of Li and Hwang has been addressed above.
Claim 14 recites “…grouping the plurality of abstracted raw data into one or more groups … based at least in part on at least one of one or more entities associated with previously determined sources of unsolicited or unwanted communications …, one or more groups of entities associated with previously determined sources …, or one or more characteristics of unsolicited or unwanted communications …”
This limitation is recited in the alternative (“at least one of”). Li discloses associating stored fingerprints with caller types and identifiers, including maintaining datasets classified by caller type (“spammers, robocallers, or known legitimate callers”) and associating fingerprints with “phone number, speaker voice fingerprints, or other identifiers associated with those callers” (¶¶[0011], [0016]). Li thus groups abstracted data by characteristics of the communications and by associated caller identity.
To the extent the combination of Li and Hwang does not expressly disclose grouping based on entities associated with previously determined sources of unwanted communications, Gallagher discloses this. Gallagher identifies and groups originating numbers associated with known robocall sources, maintaining data stores of “potential robocallers” and originating numbers that have initiated calls to disconnected numbers (¶¶[0023]–[0024], [0037]), and associating each with a robocall score. Gallagher thereby groups data based on entities/sources previously determined to be associated with unwanted communications.
It would have been obvious to group Li’s fingerprint dataset based on associated source entities as taught by Gallagher because doing so enables more efficient identification and handling of calls from known robocall sources - an objective shared by both Li and Gallagher (Li ¶[0020]; Gallagher ¶[0025]). One of ordinary skill would have combined these teachings with a reasonable expectation of success to improve detection accuracy and organization of the known-caller dataset. (See Li, ¶¶[0011], [0016], [0020]; Gallagher, ¶¶[0023]–[0025], [0037].)
Claim 9 is also rejected under 35 U.S.C. § 103 as being unpatentable over Li (US 2021/0136200 A1) in view of Grabowski et al. (US 2021/0092228 A1, “Grabowski”).
Claim 9 depends from claim 1 and recites “… one or more similarities between the first abstracted raw data and each of the plurality of abstracted raw data, based at least in part on comparison of the at least one of word count data, phoneme count data, inter-word timing data, voice pitch estimation data, or background noise data …”
The limitations of base claim 1 are disclosed by Li as set forth in the § 102 rejection above because Li discloses the base comparison and the “voice pitch estimation data” alternative.
Li discloses comparing the first abstracted raw data (voice fingerprint) with each of a plurality of abstracted raw data (stored fingerprints) by determining similarities based on voice biometrics. Li teaches that the voice biometrics used to generate and compare fingerprints include “volume, pitch, speaking rate, pauses between each utterance, tonal properties, etc.” (¶[0008]) and expressly recites generating fingerprints by “extracting … one or more voice biometrics for speakers,” which include “pitch” (¶[0035]). Li further discloses that the comparison of fingerprints is performed by “calculating a probability that the generated voice fingerprint matches one or more voice fingerprints stored in the … dataset” (¶[0014]) and generating “a similarity score” reflecting the degree of similarity (¶[0015]). Accordingly, Li discloses determining one or more similarities between the first abstracted raw data and each of the plurality of abstracted raw data based on voice pitch estimation data - one of the recited alternatives.
Grabowski discloses additional recited alternatives, including “background noise data”
To the extent it may be argued that a different recited alternative is (or will become required), Grabowski discloses analyzing additional acoustic characteristics of captured call audio data to detect robocalls, expressly including background noise data. Grabowski teaches extracting and analyzing, from captured call audio data, “amplitude levels of the captured call audio data, noise and distortion of the captured call audio data, pulse and transitions of the captured call audio data, waveforms of the captured call audio data, frequency characteristics of the captured call audio data, spectral characteristics of the captured audio data and time domain characteristics of the captured call audio data, to automatically identify entities, organizations, techniques, methods and equipment used for generating robocalls” (¶[0174]; see also ¶[0184]).
Grabowski further discloses that background noise is a distinctive characteristic used to identify robocalls, teaching that “actual live voice data, natural synthesized voice data, artificial synthesized voice data, recorded live voice data and/or recorded … synthesized voice data in the incoming call … is almost always generated with various sequences of recording speech patterns with varying background and effects including ambient, white, brown or pink noise” (¶[0246]). Grabowski dedicates further disclosure to the analysis of “[a]mbient noise,” “[w]hite noise,” “[p]ink noise,” and “[b]rown noise” as robocall indicators (¶[0247]).
Grabowski thus discloses determining similarities based on background noise data (and additionally on frequency and spectral characteristics relating to voice pitch), each of which is a recited alternative of claim 9.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate the background-noise and additional acoustic characteristic analysis taught by Grabowski into the fingerprint-comparison method of Li, and the combination would have yielded predictable results with a reasonable expectation of success, for at least the following reasons:
Both Li and Grabowski are directed to the same field of automatically detecting robocalls by analyzing acoustic characteristics of call audio in a content-independent manner (Li, ¶¶[0007]–[0009]; Grabowski, ¶¶[0026], [0174], [0235]). Grabowski expressly teaches that analyzing background noise and related acoustic characteristics improves the ability to “automatically identify entities, organizations, techniques, methods and equipment used for generating robocalls” (¶[0174]) and provides “additional levels of detection for robocalls” (¶[0026]).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to AHMAD F. MATAR whose telephone number is (571)272-7488. The examiner can normally be reached M-F 9 - 5:30.
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.
/AHMAD F. MATAR/
Supervisory Patent Examiner, Art Unit 2693
