US 7516190

I'll search for information about US Patent 7,516,190 in the USPTO records and any CAFC 2026 docket activity.

The initial searches returned no results. Let me try broader queries.

Let me search for CAFC 2026 appeal activity and the specific litigation involving this patent.

Summary: U.S. Patent No. 7,516,190 (US7516190B2)

Basic bibliographic data

• Title: Personal voice-based information retrieval system
• Inventor: Alexander Kurganov (Buffalo Grove, IL, US) — sole inventor
• Assignee: Parus Holdings, Inc. (Bannockburn, IL, US). The record shows the patent was originally assigned to Webley Systems, Inc. (May 2001), with security interests later granted to various entities, and Webley's name was changed to Parus Holdings, Inc. in January 2009.
• Application No.: 09/777,406
• Filing date: February 6, 2001
• Priority date: February 4, 2000 (Provisional Application 60/180,343)
• Issue date: April 7, 2009 (published application US20010054085A1 was published December 20, 2001)
• Status: Expired – Lifetime (adjusted expiration May 10, 2022)
• Classifications: H04M3/493 (voice portals/IVR), G06F16/638 (presentation of query results), G10L15/26 (speech-to-text), G10L13/00 (speech synthesis), among others.

Sources: Google Patents (patents.google.com/patent/US7516190/en, provided in full in the prompt); FreePatentsOnline (freepatentsonline.com/7516190.html).

Abstract

"The present invention relates to a system for retrieving information from a network such as the Internet. A user creates a user-defined record in a database that identifies an information source, such as a web site, containing information of interest to the user. This record identifies the location of the information source and also contains a recognition grammar based upon a speech command assigned by the user. Upon receiving the speech command from the user that is described within the recognition grammar, a network interface system accesses the information source and retrieves the information requested by the user."

Technology in plain language

The patent describes a personalized "voice browsing" system. A user, while on a computer, "clips" a specific portion of a web page they care about (e.g., the Chicago weather block on CNN.com), assigns it a pronounceable name ("Chicago weather"), and the system stores that name as a speech-recognition grammar along with the page's URL and a "content descriptor file" that knows where on the page the data sits. Later, from any telephone, the user speaks that name; a media server recognizes it, a web browsing server fetches the live page and extracts just the clipped portion (using regular expressions and named objects such as tables/buttons), and a speech synthesizer reads the result aloud. If the page has been redesigned and the data is no longer at the old location, the system searches the page for the named object to find the moved content.

Independent claims (plain-language overview)

The patent has 21 claims; the independent claims are 1, 6, and 16.

• Claim 1 (method): A setup-and-use method where (a) a computer with a speech processor is connected to the Internet and a phone; (b) a URL is provided; (c) a pre-defined portion of that web site containing the desired info is designated; (d) a named object associated with the content is identified; (e) a regular expression (search-pattern text string) is generated from the pre-defined portion and named object; (f–h) a speech command is given and converted to digital form, and the regular expression is assigned to that command; then, later (i–q), the user utters the audio command, it is converted, the matching regular expression is retrieved, the info is pulled from the pre-defined portion of the site — and if not found there, the site is searched for the named object — and the retrieved info is converted to an audio message and forwarded to the user.

• Claim 6 (system): A server connected to the Internet and a phone, with telephony hardware, a speech recognition engine, a speech synthesis engine, and a call processing system; a first instruction set identifying the pre-defined web-site portion (URL + named object); a second instruction set for generating the regular expression; a recognition grammar tied to a spoken command; a web browser with a content extraction agent, content fetcher, and content descriptor file; and the speech synthesis engine converting retrieved info into an audio message sent to the user. (Dependent claims cover landline/wireless/IP phones, LAN/WAN/Internet connections, a database, and a list of example named objects such as "weather," "stock," "flight," etc.)

• Claim 16 (method): A combined setup-and-retrieval method mirroring claims 1 and 6: providing the server with the two instruction sets (URL + named object; regular-expression generator), assigning a spoken command to a recognition grammar, later receiving and recognizing that command, retrieving the instruction sets, accessing the pre-defined portion of the site, retrieving the info when found, searching the web site when not found, converting the info to audio, and transmitting it to the user. (Dependent claims cover user-performed setup, graphical display/selection of the portion, and periodic updating.)

Litigation / CAFC 2026 docket findings

• The Google Patents record flags litigation: PTAB IPR2022-01542 (petitioner [Apple Inc.](/litigations/by-plaintiff/Apple%20Inc.); marked "Settlement") and PTAB IPR2022-01020 (marked "Not Instituted – Merits"), plus a Delaware District Court case (1:23-cv-00190, Parus Holdings Inc. v. Amazon.com, Inc.). The Stanford NPE Litigation Database also lists Parus Holdings Inc. v. Google, LLC, 3:22-cv-07830 (N.D. Cal.), for this patent. (Sources: patents.google.com/patent/US7516190/en; npe.law.stanford.edu/patent/7516190; portal.unifiedpatents.com/ptab/case/IPR2022-01542 and /IPR2022-01020.)
• A McKool Smith press release confirms the PTAB denied Google's IPR petitions on Parus's U.S. Patent Nos. 7,386,455 and 7,516,190, and notes the '705 patent (a related family member) was separately asserted. (Source: mckoolsmith.com/newsroom-pressreleases-366; tmcnet.com, Aug. 15, 2024.)
• CAFC 2026 docket: My searches of the CAFC 2026 docket did not surface any 2026 appeal specifically involving 7,516,190. The CAFC appeal 24-2381, Google LLC v. Parus Holdings, Inc., which had oral argument on June 5, 2026 and a decision issued July 13, 2026 (vacating a PTAB ruling), concerns the related U.S. Patent No. 6,721,705 ("Robust Voice Browser System and Voice Activated Device Controller"), not 7,516,190. (Sources: cafc.uscourts.gov/2026/06/05; ipwatchdog.com/2026/07/14/cafc-vacates-ptab-ruling-for-patent-owner-in-google-ipr; companyprofiles.justia.com/company/google/statecases/federal-appellate-decisions.)

Uncertainty notes

• My initial literal searches ("US patent 7516190 USPTO" and ""7516190" CAFC 2026 docket") returned empty result sets; the information above comes from the full patent text you supplied (Google Patents) plus follow-up web searches of FreePatentsOnline, Stanford NPE Litigation Database, Unified Patents, McKool Smith, and CAFC pages.
• I could not verify via a live search whether any CAFC 2026 appeal exists specifically for 7,516,190; the CAFC 2026 activity I found (case 24-2381) is tied to the family member '705 patent. If a CAFC docket entry for 7,516,190 exists, it did not surface in these searches, so treat the CAFC finding as "no specific 2026 CAFC case identified, with the closely related '705 appeal noted."
• The "Expired – Lifetime" status and adjusted expiration date (May 10, 2022) are as displayed by Google Patents; I did not independently verify them against USPTO PAIR/Patent Center.

Based on the searches I was able to complete, here is the known litigation involving U.S. Patent No. 7,516,190 ("the '190 patent"). I have listed only cases/PTAB proceedings where the '190 patent itself is identified as a patent-in-suit or challenged patent; I have excluded cases involving the related family member U.S. Patent No. 6,721,705 (e.g., CAFC appeal 24-2381, Google LLC v. Parus Holdings, Inc.; IPR2022-00279), which are not the '190 patent.

District court litigation

1. Parus Holdings Inc. v. Google, LLC

• Plaintiff: Parus Holdings, Inc.
• Defendant: Google, LLC
• Jurisdiction / case number: Originally U.S. District Court for the Western District of Texas, Waco Division, 6:21-cv-00571 (Judge Alan D. Albright); transferred to/re-filed in the U.S. District Court for the Northern District of California as 3:22-cv-07830-VC (Judge Vince Chhabria)
• Filing date: Original W.D. Tex. complaint filed June 4, 2021; N.D. Cal. docket opened December 12, 2022
• Patents asserted: Six patents, including the '190 patent (also 6,721,705; 7,881,941; 7,386,455; 8,185,402; 9,769,314)
• Outcome / status: Stayed February 7, 2023 pending final resolution (including appeals) of the instituted inter partes reviews. The stay motion expressly identified the '190 patent among the asserted patents. (Sources: PTACTS docket for 3:22-cv-07830; govinfo.gov opinion listing showing Judge Chhabria's order granting the stay (Document 101, filed 2/7/2023); Google's stay motion at Dkt. 94; npe.law.stanford.edu/patent/7516190; dockets.justia.com.) I could not confirm from these searches whether the stay has since been lifted or the case resolved — treat current status as "stayed as of February 2023; later status unverified."

2. Parus Holdings Inc. v. Amazon.com, Inc.

• Plaintiff: Parus Holdings, Inc.
• Defendant: Amazon.com, Inc.
• Jurisdiction / case number: U.S. District Court for the District of Delaware, 1:23-cv-00190
• Filing date: February 17, 2023
• Patents asserted: The '190 patent (per complaint analysis; the same case also asserts related patent 9,377,992)
• Outcome / status: Current status not confirmed in my searches. The complaint analysis notes the asserted patents had expired, which is consistent with the '190 patent's May 10, 2022 adjusted expiration. (Sources: npe.law.stanford.edu/patent/7516190; dockets.justia.com/docket/delaware/dedce/1:2023cv00190/81560; ai-lab.exparte.com/case/dct/ded/1:23-cv-00190.)

PTAB inter partes review proceedings

3. [Apple Inc.](/litigations/by-plaintiff/Apple%20Inc.) v. Parus Holdings, Inc. — IPR2022-01542

• Petitioner: Apple Inc.
• Patent owner: Parus Holdings, Inc.
• Patent challenged: '190 patent — all claims 1–21
• Filing date: September 16, 2022
• Outcome / status: Terminated – Settled; the Board issued a "Decision — Settlement Prior to Institution of Trial" on March 6, 2023, terminating the proceeding after a joint motion to terminate. (Sources: Patexia case summary; docket update listing showing "Terminated March 6, 2023, Status Terminated-Settled"; Google Patents litigation flag "Settlement".)

4. IPR2022-01020 (petitioner not directly confirmed in my searches)

• Patent challenged: '190 patent
• Outcome / status: Marked "Not Instituted – Merits" on the Google Patents litigation record for the '190 patent. A McKool Smith press release (August 2024) states the PTAB "denied Google's requests to review the claims of Parus' U.S. Patents Nos. 7,386,455 and 7,516,190," which strongly suggests this was a Google LLC petition against the '190 patent that was denied institution on the merits. I could not independently verify the petitioner's identity or the exact decision date in the searches completed, so treat the petitioner as "Google (per press release, not directly confirmed)." (Sources: patents.google.com/patent/US7516190/en litigation flag "Not Instituted - Merits"; tmcnet.com/McKool Smith press release, Aug. 15, 2024.)

Caveats / uncertainty notes

• Current status of the two district court cases (Google N.D. Cal. and Amazon D. Del.) is unverified beyond the February 2023 stay order in the Google case. No post-stay docket events, judgments, or dismissals surfaced in my searches.
• The Stanford NPE Litigation Database lists only the two district court cases above for the '190 patent. I did not find any ITC proceeding, CAFC appeal specifically on the '190 patent, or additional district court cases naming the '190 patent.
• The "Not Instituted – Merits" IPR (IPR2022-01020) is flagged on Google Patents but the petitioner attribution to Google is inference from the McKool Smith release, not a directly confirmed docket entry.

Proceedings overview

Two AIA trials are documented on U.S. Patent 7,516,190 — neither ever reached trial: IPR2022-01542 ([Apple Inc.](/litigations/by-plaintiff/Apple%20Inc.)) was settled and terminated before institution (2023-03-06), and IPR2022-01020 (Google LLC) was denied institution on the merits. That gives 0 active proceedings, 0 claims invalidated, 0 claims sustained by Final Written Decision, 1 settled (pre-institution), and 1 institution denied. Bottom line for a defendant: no claim of the '190 patent has ever been canceled by the PTAB — all 21 claims remain in force as issued — but the patent expired on 2022-05-10, and because no Final Written Decision ever issued, there is no IPR estoppel blocking anyone, including Google and Apple, from pressing prior-art grounds again.

Data-source flag: The USPTO Open Data Portal block supplied with this prompt reports no AIA trial proceedings on file for the '190 patent. That is a known ODP-ingest gap: the Google Patents record for US7516190B2 affirmatively lists two PTAB cases (IPR2022-01542, "Settlement"; IPR2022-01020, "Not Instituted – Merits"), and live web sources (Patexia, Unified Patents portal, Docket Alarm, McKool Smith) confirm both proceedings in detail. Both are real; the ODP block is stale.

IPR2022-01542 — Apple Inc. v. Parus Holdings, Inc.

• Type: Inter Partes Review
• Filed: 2022-09-16
• Status: Terminated – Settled (pre-institution). Board's 2023-03-06 decision: "Decision — Settlement Prior to Institution of Trial, 37 C.F.R. § 42.74." Plain-English gloss: Apple and Parus settled before the Board ever decided whether to institute; the case never went to trial and no Final Written Decision exists.
• Judge panel: Administrative Patent Judges James J. Mayberry (author of the termination decision), David C. McKone, and Patrick M. Boucher.
• Petition grounds: Apple challenged all 21 claims (1–21) of the '190 patent. The petition relied on prior art including Danner (US 7,219,136), Rajan (US 6,725,425), and Kiraly (US 6,088,731), plus a PHP ereg manual and an "Internet Scrapbook" reference, supported by a declaration from expert David Yarowsky (Ex. 1003–1004). The exact statutory mapping (§ 102 / § 103 per ground) was not verified from the available docket excerpts.
• Institution decision: None. The case settled before the institution decision was due. (Apple's exhibits show the IPR was coordinated with district-court litigation — WDTX-6-21-cv-00968 and WDTX-6-21-cv-00571, later transferred to the Northern District of California.)
• Final Written Decision: None issued.
• Settlement / termination: Joint motion to terminate filed 2023-03-03; Board terminated the proceeding on 2023-03-06 (Paper 9, "Settlement Prior to Institution of Trial"). Apple filed a refund request on 2023-03-13; the Board approved the refund on 2023-03-15. The settlement terms are not public in the docket excerpts reviewed (standard for § 42.74 terminations, which may be kept confidential).
• Appeal: None (nothing appealable — no FWD).
• Defensive value: Neutral-to-helpful for a defendant. Apple's settlement removed a live threat of a pro-Parus FWD, but because there is no FWD, the settlement creates no § 315(e)(2) estoppel and no collateral-estoppel findings against the '190 claims. Apple's prior-art set (Danner/Rajan/Kiraly) remains fully available to any defendant.

Sources: Docket Alarm/PTAB case page (gaeflexstaging-dot-docketupdate.appspot.com/cases/PTAB/IPR2022-01542/Apple_Inc._v._Parus_Holdings_Inc/); Patexia case summary (services.patexia.com/lawsuits/Apple-Inc-v-Parus-Holdings-Inc-id-181309); Unified Patents portal (portal.unifiedpatents.com/ptab/case/IPR2022-01542).

IPR2022-01020 — Google LLC v. Parus Holdings, Inc.

• Type: Inter Partes Review
• Filed: 2022-05-13 (petition and power of attorney filed same day)
• Status: Not Instituted – Merits (verbatim from the Google Patents litigation block). Plain-English gloss: the Board considered Google's challenge on the merits and declined to institute — no trial, no FWD. McKool Smith (Parus's counsel) publicly confirmed: "the Board denied Google's requests to review the claims of Parus' U.S. Patents Nos. 7,386,455 and 7,516,190."
• Judge panel: Not verified from available sources (the institution decision itself was not retrieved in these searches).
• Petition grounds: Google's petition challenged the '190 patent using a broad prior-art set — including Burrows (US 5,765,149), Dasan (US 5,761,662), Uppalaru (US 5,915,001), Brown (US 6,587,822), Iyer (US 6,446,098), Kirk (US 5,768,578), Kirsch (US 5,855,020), and D'hoore (US 6,085,160) — plus a claim-mapping table and a Sotera-style stipulation letter to Parus dated 2022-05-13 (Ex. 1029). Expert support was provided via a "Declaration of Stuart J. ___" and CV (Ex. 1002–1003; the truncated exhibit title mirrors the expert Stuart J. Lipoff used in Google's companion '455 IPR, IPR2022-00355, but I could not confirm the surname for this case from the excerpts).
• Institution decision: Denied — the Board found Google failed to show a reasonable likelihood of prevailing on the merits ("Not Instituted – Merits"). I could not verify the exact decision date or the panel's written reasoning from the sources retrieved; the denial is corroborated by McKool Smith's 2024-08-15 press release. (Companion proceeding: Google's parallel IPR on sibling patent 7,386,455, IPR2022-00355, was likewise denied — status "Not Instituted – Procedural," institution decision 2022-07-18.)
• Final Written Decision: None issued.
• Settlement / termination: No settlement — the case simply ended with the institution denial.
• Appeal: None. A § 314(d) institution denial is statutorily non-appealable.
• Defensive value: Helpful to Parus in the short term (the '190 claims survived a merits-based challenge), but no estoppel attaches to a denied petition. Google — and every other defendant — can raise the same references and combinations again in district court or in a new petition. An institution denial is a litigation data point, not a validity adjudication.

Sources: Patexia docket (services.patexia.com/lawsuits/Google-LLC-v-PARUS-Holdings-Inc-id-163787); McKool Smith press release 2024-08-15 (mckoolsmith.com/newsroom-pressreleases-366; tmcnet.com/usubmit/-mckool-smith-secures-ptab-victory-parus-holdings-against-/2024/08/15/10058788.htm); Unified Patents portal for companion case IPR2022-00355 (portal.unifiedpatents.com/ptab/case/IPR2022-00355).

Strategic summary

Claims status — CANCELED vs. SUSTAINED vs. UNTESTED. Zero claims of the '190 patent have been canceled. All 21 claims (independents 1, 6, 16 and dependents 2–5, 7–15, 17–21) remain exactly as issued — the Apple IPR settled before institution and the Google IPR was denied institution, so the Board never reached the merits in either case. The most important constraint is not the claims' validity posture but their life: the '190 patent is Expired – Lifetime (adjusted expiration 2022-05-10). Any current assertion is therefore limited to past damages for pre-expiration conduct (subject to the § 286 six-year lookback and laches/notice considerations); injunctive relief is off the table. The "hardened by IPR" framing does not apply here — Parus won at the institution gate, not on a Final Written Decision, and that buys no claim-level vindication.

Estoppel landscape. Because no Final Written Decision ever issued, 35 U.S.C. § 315(e)(2) estoppel never attached — not for Google (institution denied) and not for Apple (settlement pre-institution). Both petitioners, their privies, and every new defendant remain free to raise every ground already briefed (Apple's Danner/Rajan/Kiraly set; Google's Burrows/Dasan/Kirk/Kirsch/D'hoore/Uppalaru/Brown/Iyer set) plus any ground they reasonably could have raised. For a defendant facing a demand letter today, this is the most defendant-friendly feature of the record: there is no IPR estoppel trap, and there is a rich, already-identified prior-art menu. Watch the companion-family litigation for collateral-estoppel ripple effects, however — Parus learned in the sibling IPRs that Board findings in one family case can come back to bind it in another (see Akin Gump's analysis of the Board's collateral-estoppel-based rehearing grant in a later Google v. Parus case).

Pattern signals. This is a two-petitioner, coordinated assault on a single patent: Google filed 2022-05-13 and Apple filed 2022-09-16, both while Parus was pressing parallel district-court cases (Parus v. Google, N.D. Cal. 3:22-cv-07830; Parus v. Apple, transferred from W.D. Tex. to N.D. Cal.; and later Parus v. Amazon, D. Del. 1:23-cv-00190, which asserts the '190 against Alexa). Parus (McKool Smith) ran a successful institution-stage defense on both IPRs — a "deny-and-settle" pattern — but the broader Kurganov/Parus family is under sustained fire: sibling patents 7,076,431 and 9,451,084 were invalidated in IPRs and the Federal Circuit affirmed (Parus Holdings, Inc. v. Google LLC, 70 F.4th 1365 (Fed. Cir. 2023)); sibling patent 8,185,402 was held invalid under § 101 and affirmed (Fed. Cir. appeal 23-2297, decided 2025-03-06); and on 2026-07-13 the Federal Circuit vacated a PTAB FWD in Google's IPR on sibling patent 6,721,705 (CAFC appeal 24-2381, argued 2026-06-05) — so the family's IPR wins are being chipped away on appeal. No defensive aggregator (e.g., Unified Patents) appears as petitioner in either '190 IPR; both were filed by the operating companies themselves.

Recommended next steps

1. Lead with expiration, not IPR history. If you are the defendant in a demand letter or suit citing the '190 (e.g., the Delaware Amazon case, 1:23-cv-00190), the first defensive argument is that the patent expired 2022-05-10 — no injunction, damages only for pre-expiration conduct, and no ongoing royalty. No FWD exists to quote, because neither IPR produced one; the operative procedural records are Apple's termination decision of 2023-03-06 (Paper 9 in IPR2022-01542, available via the PTAB's docket at portal.unifiedpatents.com/ptab/case/IPR2022-01542) and Google's institution denial in IPR2022-01020 (portal.unifiedpatents.com/ptab/case/IPR2022-01020).

2. No estoppel — reuse the prior art freely. Because there was no FWD, § 315(e)(2) does not bar anyone from re-raising Google's and Apple's already-briefed grounds. In district court, put Burrows/Dasan/Kirk/Kirsch/D'hoore (Google's set) and Danner/Rajan/Kiraly (Apple's set) into your § 102/§ 103 invalidity contentions; if you file your own IPR, be mindful of the § 315(b) one-year bar from service and of the Board's discretion (Fintiv) while the Delaware/NDCal cases are pending.

3. Press § 101. The Federal Circuit's 2025 affirmance of § 101 invalidity on the sibling '402 patent (Parus Holdings v. Google, appeal 23-2297) — same inventor, same Feb. 4, 2000 priority family, same voice-command/retrieval architecture — gives you a roadmap for an Alice motion on the '190's "regular expression + speech command" claims. The live claim-construction battleground is the term "regular expression" (the Amazon complaint itself flags that Alexa's "intents"/"sample utterances" mapping to the claimed regex is the pivotal, and weakest, infringement theory).

4. Don't over-invest in new IPR economics. With the patent expired, a new IPR's value is confined to estoppel and damages-defense; the Board can institute on expired patents, and a FWD would then hand the petitioner § 315(e)(2) estoppel armor. But weigh that against cost and the strong likelihood that the district court will reach validity first.

5. Monitor the family appeals. The 2026-07-13 CAFC vacatur in appeal 24-2381 (Google v. Parus, '705 patent) signals that the Federal Circuit is actively reviewing this family's PTAB outcomes — a favorable (to defendants) development that could ripple into claim-construction and obviousness positions on the '190.

Caveat: I could not independently verify the institution-decision date or panel for IPR2022-01020, nor the exact statutory grounds asserted claim-by-claim in either petition, from the sources retrieved; the "Not Instituted – Merits" status and the denial itself are confirmed by the Google Patents record and McKool Smith's press release. No CAFC appeal exists for either '190 IPR because neither produced a final written decision.

Inventors

• Alexander Kurganov (sole inventor). Employer at time of filing: not stated on the face of the patent, but the record is telling: the application (09/777,406) was filed 2001-02-06 and the very first recorded conveyance (executed 2001-05-14) assigns the inventor's interest to Webley Systems, Inc. — the company whose internal systems ("a database 300 designed by Webley Systems Incorporated") are described throughout the specification. Kurganov was the founder/principal of Webley Systems. There is no evidence of any other named inventor, and no unusual pattern of post-filing inventor departures is visible from the assignment record.

Original assignee

• Webley Systems, Inc. (later renamed Webley Corporation dba Webley Systems, Inc., and then renamed Parus Holdings, Inc. on 2009-01-23). Google Patents displays "Original Assignee: Parus Holdings Inc" because of that rename, but the recorded chain of title shows the patent originated with Webley Systems.
• Product status: Webley Systems was a voice-portal / unified-communications company (late 1990s–early 2000s); the patent describes systems Webley itself built, so the claims were embodied in Webley's own voice-browsing product line at the time of filing. The operating business, however, is long defunct.
• Current status: The entity survives only as Parus Holdings, Inc., which since at least 2021 has acted exclusively as a patent-assertion plaintiff (suits against Google and Amazon, multiple PTAB IPRs defending the portfolio). No product business is evident from the litigation record.

Assignment timeline

I was not able to retrieve reel/frame numbers or correspondent-of-record data for this patent from the USPTO Assignment Center via the search tools available to me — the USPTO Assignment Center is at https://assignmentcenter.uspto.gov/ and should be consulted to fill in reel/frame and correspondent fields. I am not fabricating those identifiers. The chronological record below is taken from the Google Patents legal-events display (authoritative patent text supplied), which is consistent with USPTO-recorded conveyances (assignment, security agreements, change of name) and identifies the parties and dates, but does not include reel/frame or correspondent names.

• 2001-05-14 (recorded) — Conveyance: Assignment of Assignor's Interest; Assignor: Alexander Kurganov; Assignee: Webley Systems, Inc. (Reel/frame and correspondent: not retrievable in my search; verify at USPTO Assignment Center.) Context: inventor-to-company assignment at filing; standard first leg of title.
• 2001-09-24 (recorded) — Conveyance: Security Agreement; Assignor: Webley Systems, Inc.; Assignees: R. Scot Thomas, World Wide Ventures, LLC, Vail Systems, Inc. (Reel/frame: not retrievable.) Context: venture-lending/lien grant against the patent portfolio — the first of several debt financings, not a title transfer.
• 2002-02-28 (recorded) — Conveyance: Security Agreement; Assignor: Webley Systems, Inc.; Assignee: Global CK Ventures, LLC. (Reel/frame: not retrievable.) Context: additional secured financing against the portfolio.
• 2002-02-28 (recorded) — Conveyance: Security Interest; Assignor: Webley Systems, Inc.; Assignee: Global CK Ventures, LLC. (Reel/frame: not retrievable.) Context: related security-interest recordal on the same date.
• 2002-02-28 (recorded) — Conveyance: Security Agreement; Assignor: Webley Systems, Inc.; Assignee: World Wide Ventures, LLC. (Reel/frame: not retrievable.) Context: additional lien holder recorded.
• 2003-08-22 (recorded) — Conveyance: Security Interest; Assignor: Webley Systems, Inc.; Assignee: Global CK Ventures, LLC. (Reel/frame: not retrievable.) Context: follow-on security-interest recordal.
• 2003-08-22 (recorded) — Conveyance: Security Interest; Assignor: Webley Systems, Inc.; Assignee: WWIS LLC. (Reel/frame: not retrievable.) Context: new lender/security holder added to the portfolio.
• 2009-01-23 (recorded) — Conveyance: Change of Name; Assignor: Webley Corporation dba Webley Systems, Inc.; Assignee: Parus Holdings, Inc. (Reel/frame: not retrievable.) Context: pure name change — the operating company reorganized into the holdings entity that later became the asserter.
• 2013-03-18 (recorded) — Conveyance: Security Agreement; Assignor: Parus Holdings, Inc.; Assignee: WWIS, LLC. (Reel/frame: not retrievable.) Context: a security interest granted nearly a decade after the others — the last recorded event on the file, eight years before the first infringement suit.

No further assignments appear on the Google Patents legal-events record after 2013. The title never left the Webley→Parus family; the LLCs in the chain (World Wide Ventures, Global CK Ventures, WWIS LLC, and individuals R. Scot Thomas / Vail Systems) hold security interests/liens, not ownership.

Timeline diagram

timeline
    title Ownership of US 7516190
    2001 : Filed by Kurganov
         : Assigned to Webley Systems
         : Security agreement to Thomas WWV Vail
    2002 : Security interests to Global CK
         : Security interest to World Wide Ventures
    2003 : Security interest to Global CK
         : Security interest to WWIS LLC
    2009 : Name change to Parus Holdings
    2013 : Security agreement to WWIS LLC
    2021 : First infringement suit filed

NPE / troll-pattern signals

1. Shell-entity transfer — not present as a recorded transfer / unclear as to function. The recorded chain contains no transfer to a separate licensing LLC: the current owner, Parus Holdings, Inc., arose by Change of Name (2009-01-23) from Webley Corporation, not by a transfer to a new shell. The "Holdings" suffix and total absence of any product activity in the litigation record support non-practicing status, but I cannot cite a registered-agent address or single-purpose LLC formation from my retrievable sources, so I flag this as naming-plus-litigation inference only.

2. Known asserter in the chain — present (strong). Parus Holdings, Inc. is the current assignee (via the 2009-01-23 change of name) and is listed as the patent asserter for this exact patent in the Stanford NPE Litigation Database: Parus Holdings Inc. v. Amazon.com, Inc., 1:23-cv-00190 (D. Del., filed 2023-02-17) and Parus Holdings Inc. v. Google, LLC, 3:22-cv-07830 (N.D. Cal.; originally W.D. Tex. 6:21-cv-00571, filed 2021-06-04). The patent was also challenged in IPR2022-01542 (Apple; settled before institution, terminated 2023-03-06) and IPR2022-01020 (denied institution on the merits). This is concrete, citable evidence of a high-frequency assertion program, not a naming inference.

3. Repeat correspondent across the chain — unclear. I could not retrieve correspondent-of-record data for any reel/frame entry in my searches. This is a data gap, not a negative finding.

4. Cascading transfers — not present. The 2001–2003 and 2013 entries are all security agreements/security interests (liens), not ownership transfers; there is no chain of chained LLC assignments.

5. Pre-litigation transfer — not present. The last recorded event (2013-03-18 security agreement to WWIS LLC) predates the first infringement suit (June 2021) by more than eight years. The 2009 rename to Parus Holdings predates the suits by a decade.

6. Bankruptcy fire-sale — unclear / not established. The cluster of security agreements in 2001–2003 evidences severe financial distress at Webley (multiple lenders taking liens), but my searches surfaced no Chapter 7/11 filing or court-supervised patent sale. Do not treat distress as bankruptcy without a docket.

7. Privateering — not present. There is no operating company transferring the patent to a third-party NPE to sue competitors on its behalf; Parus is the renamed successor of the original company, and its targets (Google, Amazon) are not competitors of a defunct voice-portal operator.

8. Defensive aggregator (anti-NPE) — not present. The chain terminates at Parus Holdings, an asserting entity; no RPX/AST/LOT/Unified/OIN involvement in the title.

Verdict

NPE — moderate confidence.

The decisive evidence is the known- asserter signal: the current assignee, Parus Holdings, Inc. (formed by the 2009-01-23 change of name from Webley Corporation), is listed in the Stanford NPE Litigation Database as the asserter of this very patent in two district court actions (Google N.D. Cal. 3:22-cv-07830, filed 2021; Amazon D. Del. 1:23-cv-00190, filed 2023), and the patent — which expired 2022-05-10 — is being asserted for past damages against non-competitors, consistent with a monetization vehicle rather than an operating company. I stopped short of "high confidence" because I could not retrieve the USPTO reel/frame and correspondent-of-record data to confirm the full chain and repeat-filer patterns, and I could not independently verify Parus's non-practicing status beyond the litigation record. Verify the recorded conveyances at the USPTO Assignment Center: https://assignmentcenter.uspto.gov/ (search by patent number 7516190; also indexed at https://assignment.uspto.gov/patent/index.html).

Prior Art Analysis: U.S. Patent No. 7,516,190 ("Personal Voice-Based Information Retrieval System")

1. USPTO database verification for patent number 7516190

Searches of the USPTO's public databases (via FreePatentsOnline, Google Patents, and uspto.report mirrors of the USPTO grant) confirm the specific patent US7516190B2, application 09/777,406, filed 2001-02-06, claiming priority to provisional 60/180,343 filed 2000-02-04, issued 2009-04-07 to inventor Alexander Kurganov, assigned to Parus Holdings, Inc. (formerly Webley Systems, Inc.). No similarly numbered patents were substituted; the analysis below is specific to 7516190.

Critical date for § 102 analysis: Because the '190 patent validly claims priority to the February 4, 2000 provisional application, a reference is prior art under pre-AIA § 102(a) if it was patented or published before February 4, 2000, and under pre-AIA § 102(e) if its earliest U.S. filing date precedes February 4, 2000 (even if published later). References first filed after February 4, 2000 are not § 102 prior art (they could only support § 103 combinations).

2. Claim-element framework used for the mapping

The independent claims are 1 (method), 6 (system), and 16 (method). Their limiting combination is:

• (A) a computer/server with a speech processor connected to the Internet and to a phone (claim 1(a); 6; 16);
• (B) a URL identifying a pre-selected web site (1(b); 6; 16);
• (C) designation of a pre-defined portion of that web site containing the desired information (1(c); 6; 16);
• (D) identification of a named object associated with the content (1(d); 6; 16);
• (E) generation of a regular expression (search-pattern text string) based on the pre-defined portion and named object (1(e); 6; 16);
• (F) a user-assigned speech command mapped to a recognition grammar (1(f)–(h); 6; 16);
• (G) later utterance → recognition → retrieval of the stored record (1(i)–(l); 6; 16);
• (H) retrieval of the information from the pre-defined portion, and if not found, searching the site for the named object (1(m)–(n); 6; 16);
• (I) speech-synthesis conversion and audio delivery to the user (1(o)–(q); 6; 16).

For each reference below, "potentially anticipates claim X" means the reference, read in light of the '190 specification and with reasonable claim interpretation, appears to disclose the recited combination (or all but a minor element that could be argued inherent). Where a reference covers only a subset of elements, I say so explicitly.

3. Tier 1 — Most relevant prior art (detailed analysis)

3.1 US5953392A — Netphonic Communications, Inc., "Method and apparatus for telephonically accessing and navigating the internet"

• Dates: Filed 1996-03-01; granted 1999-09-14 (prior art under § 102(a)/(b)).
• Description: A caller uses a telephone to access Internet content. The system receives a URL (via DTMF or voice), fetches the HTML page over HTTP, converts it to a voice-navigable structure (H-tree), and reads text aloud using synthesized or recorded voice; hyperlinks are announced and selectable. Supports fax/email/postal delivery of pages.
• Claim mapping: Discloses elements (A) (phone + computer network), (B) (URL fetch), (G) (request → retrieval), (I) (voice synthesis output). It does not disclose user designation of a pre-defined page portion (C), a named object (D), a generated regular expression (E), or the fallback named-object search (H). Potentially anticipates claims 1, 16 only under an aggressive reading that treats page-level retrieval as the "pre-defined portion"; more realistically it is the strongest single reference against the telephony/voice-browsing preamble elements of claims 1, 6, 16 and anticipates claim 2/20-type "periodically updated" and claim 7–9/12 phone-type limitations only in combination with other art.

3.2 US5884262A — Bell Atlantic Network Services, Inc., "Computer network audio access and conversion system"

• Dates: Filed 1996-03-28; granted 1999-03-16 (§ 102(a)/(b)).
• Description: A network audio access system that lets telephone callers request (by voice or keypad) information from computer-network sources (including web pages); the system fetches the document and converts text to speech for telephone delivery.
• Claim mapping: Discloses (A), (B), (G), (I). Does not disclose (C)/(D)/(E)/(H). Potentially anticipates the telephone-audio-retrieval framework of claims 1, 6, 16; lacks the clipping/named-object/regular-expression elements.

3.3 US6269336B1 — Motorola, Inc. (Ladd et al.), "Voice browser for interactive services and methods thereof"

• Dates: Filed 1998-07-24; granted 2001-07-31 (§ 102(e) prior art; filed ~18 months before the critical date).
• Description: A voice browser enabling users to access web information sources (news, designated city weather, traffic, stock quotes) via speech over any network. A VRU server compares speech against a vocabulary/grammar; a markup language (VoxML) defines dialogs with prompts, input elements, and a barge-in attribute; content is identified by URL/URN and read aloud via TTS.
• Claim mapping: Discloses (A) (communication node + telephony + speech recognition + synthesis), (B) (URL-identified information source), (F) (grammar-based speech command recognition), (G), (I), and the "barge-in" feature of the '190 specification. Does not disclose user designation of a pre-defined portion (C), named object (D), regular expression generation (E), or the moved-content fallback (H). Potentially anticipates claims 6 and 16's system/method framework; the closest reference to the '190's media-server + grammar architecture. Claim 6's "recognition grammar corresponding to each instruction set" and claim 1's speech-command-to-grammar mapping are strongly present.

3.4 US6539359B1 — Motorola, Inc. (Ladd et al.), "Markup language for interactive services and methods thereof"

• Dates: Filed (priority) 1998-10-01; granted 2003-03-25 (§ 102(e)).
• Description: Companion to 6269336; defines a markup language document with dialog/prompt/input/barge-in elements used to build voice applications that retrieve web content.
• Claim mapping: Same coverage as 3.3; potentially anticipates claims 6 and 16's instruction-set/markup-driven retrieval architecture, absent elements (C)/(D)/(E)/(H).

3.5 US5819220A — Hewlett-Packard Company, "Web triggered word set boosting for speech interfaces to the world wide web"

• Dates: Filed 1996-09-30; granted 1998-10-06 (§ 102(a)/(b)).
• Description: A speech interface to the WWW in which the vocabulary/grammar available to the recognizer is dynamically boosted using words/tokens extracted from a fetched web page, improving recognition of page-specific content (e.g., links) spoken by the user.
• Claim mapping: Discloses (A), (B), (F) (dynamic grammar tied to web content), (G). Relevant to the '190's concept of page-derived recognition content, but lacks (C)/(D)/(E)/(H). Potentially anticipates claim 1(f)–(h)/16's speech-command-to-grammar assignment in a web context; not the full combination.

3.6 US6101472A — IBM Corp., "Data processing system and method for navigating a network using a voice command"

• Dates: Filed 1997-04-16; granted 2000-08-08 (§ 102(a)/(b) — published after critical date but filed before; also § 102(e)).
• Description: Voice-command navigation of a network (web) browser; spoken commands are recognized and translated into navigation actions; retrieved pages are audibly presented.
• Claim mapping: (A), (B), (F), (G), (I). Lacks (C)/(D)/(E)/(H). Potentially anticipates claim 1/16's voice-command web retrieval and claim 6's speech-recognition-engine-plus-browser elements.

3.7 US5884266A — Motorola, Inc., "Audio interface for document based information resource navigation and method therefor"

• Dates: Filed 1997-04-02; granted 1999-03-16 (§ 102(a)/(b)).
• Description: Audio/document interface that navigates document-based information resources (web documents) using speech recognition and TTS; documents are parsed into structural elements for audio navigation.
• Claim mapping: (A), (B), (F), (G), (I), and arguably (C) (selecting/parsing a portion of a document for audio rendering). Lacks (D)/(E)/(H). Potentially anticipates claims 1, 6, 16's core voice-browsing steps; the document-parsing disclosure is the closest to "pre-defined portion."

3.8 US5915001A — Vois Corporation, "System and method for providing and using universally accessible voice and speech data files"

• Dates: Filed 1996-11-14; granted 1999-06-22 (§ 102(a)/(b)).
• Description: Universal voice/speech data files (analogous to web files) that can be accessed by telephone; voice commands navigate and retrieve audio content from a network.
• Claim mapping: (A), (B) (voice-file address), (F), (G), (I). Lacks (C)/(D)/(E)/(H). Potentially anticipates claims 1, 16's phone-based retrieval method and claim 6's system elements.

3.9 US6018710A — Siemens Corporate Research, "Web-based interactive radio environment: WIRE"

• Dates: Filed 1996-12-13; granted 2000-01-25 (§ 102(a)/(b)).
• Description: A web-based "interactive radio" where users select web information channels (news, weather, etc.) and the content is converted to audio streams; user preferences personalize the information delivered.
• Claim mapping: (A), (B), (C) (user-selected channels = pre-defined content), (I). Lacks (D)/(E)/(F)/(H) as claimed. Potentially anticipates claims 1, 6, 16's user-customized audio delivery; the personalization is closest to the '190's "user-defined search" concept.

3.10 US6349132B1 — Talk2 Technology, Inc., "Voice interface for electronic documents"

• Dates: Filed 1999-12-16; granted 2002-02-19 (§ 102(e); filed before critical date).
• Description: A voice interface for electronic documents (web pages) where a user speaks to select/access content; the system maps spoken input to document elements and reads selected content aloud.
• Claim mapping: (A), (B), (F), (G), (I), and arguably (C)/(D) (selecting document elements/content). Lacks (E)/(H). Potentially anticipates claims 1, 6, 16's voice-selection of document content; a strong § 102(e) reference because it was filed ~6 weeks before the '190 provisional.

3.11 US6327572B1 — Talk2 Technologies, Inc., "Viral marketing for voice-accessible information service"

• Dates: Filed 1999-10-13; granted 2001-12-04 (§ 102(e)).
• Description: Voice-accessible information service with user registration of personal information sources, user-selected content ("favorites"), and telephone voice access with TTS delivery.
• Claim mapping: (A), (B), (C) (user-selected favorites), (F), (G), (I). Lacks (D)/(E)/(H). Potentially anticipates claims 1, 6, 16's user-defined-search concept.

3.12 US20010032234A1 — Summers, "Mapping an internet document to be accessed over a telephone system"

• Dates: Filed 1999-12-16; published 2001-10-18 (§ 102(e); filed before critical date).
• Description: Maps Internet documents (URLs) to telephone-accessible audio renditions; a user accesses a web document by phone and hears it read aloud; includes document-to-voice mapping metadata.
• Claim mapping: (A), (B), (F), (G), (I); arguably (C) (mapped portions of the document). Lacks (D)/(E)/(H). Potentially anticipates claims 1, 16's setup-and-retrieve method (the mapping step parallels the '190's clipping step).

3.13 US6446076B1 — Accenture LLP, "Voice interactive web-based agent system responsive to a user location for prioritizing and formatting information"

• Dates: Filed 1998-11-12; granted 2002-09-03 (§ 102(e)).
• Description: Voice-interactive web agent that retrieves web information based on spoken user requests, formats/prioritizes it, and reads it back over a telephone.
• Claim mapping: (A), (B), (F), (G), (I). Lacks (C)/(D)/(E)/(H). Potentially anticipates claims 1, 6, 16's agent-based retrieval framework.

3.14 US6456699B1 — AT&T Corp., "Web-based generation of telephony-based interactive voice response applications"

• Dates: Filed 1998-11-30; granted 2002-09-24 (§ 102(e)).
• Description: Generating telephony IVR applications from web-based definitions; users interact by voice with applications that fetch web content.
• Claim mapping: (A), (B), (F), (G), (I). Potentially anticipates claims 6 and 16's IVR/call-processing architecture (the '190's media server elements).

3.15 US6532444B1 — One Voice Technologies (Weber), "Network interactive user interface using speech recognition and natural language processing"

• Dates: Filed (priority) 1998-09-09; granted 2003-03-11 (§ 102(e)).
• Description: Network user interface using speech recognition and natural language processing to access web content; voice commands are recognized against grammars and used to retrieve/control networked applications.
• Claim mapping: (A), (B), (F), (G), (I). Potentially anticipates claims 1, 6, 16's speech-recognition-driven network retrieval.

3.16 US7050977B1 — IBM Corp. (Bennett), "Speech-enabled server for internet website and method"

• Dates: Priority 1999-11-11; granted 2006-05-23 (§ 102(e); filed before critical date).
• Description: A speech-enabled web server that receives spoken user input over a network, performs speech recognition, queries web content, and returns audio (TTS) responses — effectively a voice browser server.
• Claim mapping: (A), (B), (F), (G), (I). Lacks (C)/(D)/(E)/(H). Potentially anticipates claims 6 and 16's server-side speech-processing architecture.

3.17 US20020006126A1 — Johnson, "Methods and systems for accessing information from an information source"

• Dates: Priority 1998-07-24; published 2002-01-17 (§ 102(e); same family as 6269336).
• Description: Methods/systems for accessing information from an information source (web site) via a voice browser using grammars and markup.
• Claim mapping: Same as 3.3; potentially anticipates claims 6, 16's grammar-based voice-browser retrieval.

3.18 US20010048676A1 — Jimenez, "Methods and apparatus for executing an audio attachment using an audio web retrieval telephone system"

• Dates: Priority 2000-01-07; published 2001-12-06 (§ 102(e) if the U.S. filing predates 2000-02-04 — verify in Patent Center; the listed priority date is before the critical date).
• Description: An audio web-retrieval telephone system where users retrieve audio attachments/web audio content by phone.
• Claim mapping: (A), (B), (G), (I). Potentially anticipates claims 1, 16's telephone audio retrieval; weaker on the grammar/clipping elements.

3.19 US6285745B1 — Bell Atlantic Network Services, Inc., "Analog terminal internet access"

• Dates: Filed 1994-12-05; granted 2001-09-04 (§ 102(a)/(b)).
• Description: Providing Internet access (including web content) to analog telephone terminals (e.g., screen phones) with audio conversion.
• Claim mapping: (A), (B), (I). Potentially anticipates the phone-to-Internet gateway elements of claims 1, 6, 16.

3.20 US6243373B1 — Telecom Internet Ltd., "Method and apparatus for implementing a computer network/internet telephone system"

• Dates: Filed 1995-11-01; granted 2001-06-05 (§ 102(a)/(b)).
• Description: Internet telephone system enabling telephone access to computer-network data with voice output.
• Claim mapping: (A), (B), (I). Potentially anticipates the telephony/Internet integration elements of claims 1, 6, 16.

3.21 US6233318B1 — Comverse Network Systems, "System for accessing multimedia mailboxes and messages over the internet and via telephone"

• Dates: Filed 1996-11-05; granted 2001-05-15 (§ 102(a)/(b)).
• Description: Unified access to multimedia messages/mailboxes via both Internet and telephone with speech recognition and TTS.
• Claim mapping: (A), (F), (G), (I). Potentially anticipates claims 1, 6, 16's multi-modal (phone + network) access architecture.

3.22 US6252944B1 — Estech Systems, Inc., "Telephone call/voice processing system"

• Dates: Filed 1997-06-11; granted 2001-06-26 (§ 102(a)/(b)).
• Description: Telephone voice processing system with speech recognition, TTS, and call processing for retrieving/announcing information.
• Claim mapping: (A), (F), (G), (I). Potentially anticipates claim 6's telephony-hardware + recognition + synthesis + call-processing system elements.

3.23 US6353661B1 — Bailey, III, "Network and communication access systems"

• Dates: Filed 1997-12-18; granted 2002-03-05 (§ 102(a)/(b)).
• Description: Network/communication access systems providing voice access to networked information.
• Claim mapping: (A), (B), (I). Potentially anticipates claims 1, 6, 16's network-access preamble elements.

3.24 US6366578B1 — Verticle Networks, Inc., "Systems and methods for multiple mode voice and data communications using intelligently bridged TDM and packet buses"

• Dates: Filed 1998-04-03; granted 2002-04-02 (§ 102(e)).
• Description: Multi-mode voice/data communications platform bridging telephony (TDM) and packet networks (e.g., IP), enabling voice access to data services.
• Claim mapping: (A). Potentially anticipates claim 6's telephony/IP hardware elements.

3.25 US6208638B1 — J2 Global Communications, "Method and apparatus for transmission and retrieval of facsimile and audio messages over a circuit or packet switched network"

• Dates: Filed 1997-04-01; granted 2001-03-27 (§ 102(a)/(b)).
• Description: Retrieval of audio/fax messages over circuit or packet networks by phone.
• Claim mapping: (A), (G), (I). Potentially anticipates claims 1, 16's telephone message-retrieval elements.

3.26 US5974413A — Activeword Systems, Inc., "Semantic user interface"

• Dates: Filed 1997-07-03; granted 1999-10-26 (§ 102(a)/(b)).
• Description: Semantic user interface allowing users to issue commands (including spoken) to access application content semantically.
• Claim mapping: (F), (G). Potentially anticipates claims 1, 16's spoken-command-to-content mapping.

3.27 US5799065A — Matsushita Electric Industrial Co., "Call routing device employing continuous speech"

• Dates: Filed 1996-05-06; granted 1998-08-25 (§ 102(a)/(b)).
• Description: Call routing using continuous speech recognition.
• Claim mapping: (F). Potentially anticipates claims 1, 16's speech-command recognition step and claim 6's speech-recognition engine.

3.28 US5719921A — Nynex Science & Technology, "Methods and apparatus for activating telephone services in response to speech"

• Dates: Filed 1996-02-29; granted 1998-02-17 (§ 102(a)/(b)).
• Description: Activating telephone services (including information services) by spoken commands recognized against stored grammars.
• Claim mapping: (F), (G). Potentially anticipates claims 1, 16's speech-activated service invocation and claim 6's grammar/recognition elements.

3.29 US5721908A — IBM Corp., "Computer network for WWW server data access over internet"

• Dates: Filed 1995-06-07; granted 1998-02-24 (§ 102(a)/(b)).
• Description: WWW server data access over the Internet with audio/voice conversion of web content.
• Claim mapping: (A), (B), (I). Potentially anticipates claims 1, 6, 16's web-fetch-and-audio-output elements.

3.30 US5699486A — Canon Information Systems, Inc., "System for speaking hypertext documents such as computerized help files"

• Dates: Filed 1993-11-24; granted 1997-12-16 (§ 102(a)/(b)).
• Description: System for audibly reading hypertext documents, parsing document structure for spoken output.
• Claim mapping: (B), (I), arguably (C). Potentially anticipates claims 1, 16's document-to-speech element.

4. Tier 2 — Relevant to specific claim elements (moderate detail)

5. Tier 3 — Background/telephony/voicemail references (brief; individually unlikely to anticipate)

These 1980s–1990s references are cited for background concepts (unified messaging, voice control, call processing, fax store-and-forward, IVR). None discloses web-content clipping, named objects, or regular-expression generation, so none alone anticipates claims 1, 6, or 16; they are relevant only to § 103 combinations on elements (A)/(F)/(G)/(I).

• US4313035A (Bell Labs, 1982) — person locator service; US4585906A (VMX, 1986) — electronic audio communication with user-controlled message address; US4596900A (1986) — tone-operated control device; US4696028A (Dytel, 1987) — PBX intercept/attendant bypass; US4761807A (VMX, 1988) — audio communications with voice authentication; US4776016A (1988) — voice control system; US4809321A / US4935958A / US4922526A / US4850012A / US4955047A / US4975941A / US5020095A / US5027384A / US5029196A / US5099509A / US5109405A / US5166974A (Dytel family, 1989–1992) — automated call processing, routing, screening, attendant; US4837798A (AT&T, 1989) — unified messaging; US4873719A (1989) — calling number delivery; US4907079A (1990) — home entertainment monitoring; US4933966A (1990) — automated collect call; US4953204A (AT&T, 1990) — multilocation queuing; US4972462A (Hitachi, 1990) — multimedia mail; US4994926A / US5459584A / US5404231A / US5555100A (Audiofax/F-Mail, 1991–1996) — fax store-and-forward; US5086385A (1992) — home automation; US5131024A (1992) — proactive call services; US5195086A / US5243645A / US5384771A / US5408526A (AT&T, 1993–1995) — multimedia conferencing/call forwarding/call configuration; US5249219A (Dytel, 1993) — voice store-and-forward; US5263084A / US5329578A (Nortel, 1993–1994) — caller ID with call waiting / personal communication service; US5291479A (1994) — modular programmable telecom system; US5307399A / US5657376A (Glenayre, 1994–1997) — paging/voice messaging architecture; US5309504A (1994) — attendant position identification; US5325421A (AT&T, 1994) — voice-directed communications platform; US5327529A (Geoworks, 1994) — UI design process; US5327486A (Bellcore, 1994) — telecom service management; US5333266A (IBM, 1994) — message handling; US5355403A (1994) — billing-controlled call bridging; US5375161A / US5752191A (Accessline, 1994–1998) — telephone control/branch routing; US5436963A (IBM, 1995) — telephone answering; US5463684A (Microlog, 1995) — call transfer without PBX; US5475791A (1995) — spoken-word recognition amid noise; US5497373A (Ericsson, 1996) — multi-media interface; US5499288A / US5517558A / US5659597A (Voice Control Systems, 1996–1997) — voice recognition/verification for telephone services; WO1996009710A1 (Octel, 1996) — network multimedia communications/directory; US5566236A (1996) — re-establishing disconnected calls; US5603031A / US5611031A / US5675811A / US5689669A / US5692187A / US5787298A / US5793993A / US5812796A / US5819306A (General Magic, 1997–1998) — distributed computing/UI/serial bus; US5608786A (1997) — unified messaging; US5652789A / US6021181A (Wildfire, 1997–2000) — network-based knowledgeable assistant / voice mail handling; US5675507A (1997) — message storage/delivery; WO1997037481A1 (Nortel, 1997) — speech-recognition vocabulary reduction; WO1998023058A3 (Premiere, 1998) — integrated messaging management; US5742905A (Bellcore, 1998) — personal communications internetworking; US5809282A (1998) — network simulation; US5867495A / US5867494A / US5999525A (MCI, 1999) — hybrid-network communications/billing/videotelephony; US5873080A (IBM, 1999) — multiple search engines for multimedia; US5881134A (1999) — intelligent call processing platform; US5890123A (Lucent, 1999) — voice-controlled video screen display; US6012088A (IBM, 2000) — automatic configuration for Internet access; US6014437A (IBM, 2000) — multi-service platform for telephone networks; US6031904A (Nortel, 2000) — service order mechanism; US6038305A (Bell Atlantic, 2000) — personal dial tone with personalized caller ID; US6078580A (IBM, 2000) — operator status server for voice over data; US6081518A (Andersen, 2000) — cross-location registration; US6091808A (Nortel, 2000) — telephone call control/information; US6104803A (Alcatel, 2000) — intelligent service peripheral; US6115742A (AT&T, 2000) — secure metering over a communications network; GB2317782B (Matsushita, 2001) — voice dialling server for PBX; US6505163B1 (Bellsouth, 2003) — automatic recall with ASR (filed 2000-08-09 — after the critical date; not § 102 prior art).

6. References cited in the '190 patent that are NOT § 102 prior art (filed after 2000-02-04)

These appear on the face of the '190 patent but cannot anticipate under § 102 because their earliest U.S. filing dates post-date the February 4, 2000 critical date:

• US6434529B1 (Sun Microsystems; filed 2000-02-16) — speech-recognition-grammar object references.
• US6477420B1 (Medtronic; filed 2001-04-27) — implantable pacemaker rate control (unrelated field; cited for background).
• US6505163B1 (Bellsouth; filed 2000-08-09) — automatic recall with ASR.
• (References in Tier 2 whose filing dates I could not verify, e.g., US6618726, US6594348, US6687341, US6732142, US6718015, US6964012, US6965864, US6823370, US6922733, US6941273, US6888929, US6996609, US6546393, US6529948, US6665640, US6771743, must be checked in USPTO Patent Center: if any was filed before 2000-02-04, it becomes § 102(e) art — several of these are the most factually relevant to the "pre-defined portion"/"select content" limitations, so this verification materially affects the anticipation analysis.)

7. Bottom line

1. USPTO record confirmed: Patent 7516190 = "Personal voice-based information retrieval system" (Kurganov; Parus Holdings; filed 2001-02-06; priority 2000-02-04; granted 2009-04-07). No similar-number substitutions.
2. Most relevant prior art: The strongest § 102 candidates against the independent claims are US6269336B1 and its family member US6539359B1 (voice browser + recognition grammars + markup-driven retrieval; § 102(e)), US6349132B1 and US6327572B1 (Talk2; voice interface to electronic documents; § 102(e), filed weeks before the critical date), US5953392A, US5884262A, US5819220A, US6101472A, US5884266A, US5915001A, and US6018710A (§ 102(a)/(b) voice/web/telephone retrieval), plus US7050977B1, US6446076B1, US6456699B1, US6532444B1, US20010032234A1, and US20020006126A1 (§ 102(e)).
3. Anticipation caveat: No single reference I examined appears to disclose the entire claimed combination — in particular the trio of user-designated pre-defined portion (C) + named object (D) + regular-expression generation (E) and the fallback search for the named object when content moves (H). The closest references on those specific limitations are US6823370B1 ("retrieving select web content"), US6687341B1 (customized content via telephone), US6349132B1, and US20010032234A1 — but their filing dates (and hence § 102(e) status) need to be verified in USPTO Patent Center, which I could not complete within the search limits. If a challenger establishes that any of these was filed before February 4, 2000, that reference becomes the primary anticipation threat to claims 1, 6, and/or 16; otherwise the '190's combination likely survives § 102 and the fight shifts to § 103 obviousness.
4. Uncertainty note: Filing dates for approximately 17 Tier-2 references and the petitioner identity for IPR2022-01020 remain unverified; treat those items as requiring Patent Center confirmation rather than established fact.

Obviousness Analysis — U.S. Patent No. 7,516,190 (the "'190 patent") under 35 U.S.C. § 103

1. Scope and framework

This analysis addresses whether the three independent claims — claim 1 (setup-and-retrieval method), claim 6 (system), and claim 16 (combined method) — and their dependent claims (2–5, 7–15, 17–21) would have been obvious to a person of ordinary skill in the art ("PHOSITA") as of the critical date (February 4, 2000, the priority date of provisional application 60/180,343, to which the '190 patent claims priority).

The analysis applies the Graham factors (scope and content of the prior art; differences between the prior art and the claims; level of ordinary skill; secondary considerations) as refined by KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007) — i.e., a flexible, common-sense obviousness inquiry in which known elements can be combined "according to known methods" with a "reasonable expectation of success," and in which the motivation may be found in the nature of the problem, the teachings of the references, or the general knowledge of the art.

2. The person of ordinary skill in the art (PHOSITA)

A PHOSITA in early 2000 would hold a B.S. in computer science, computer engineering, or electrical engineering (or equivalent experience) with 2–4 years of practical experience in at least two of: (i) telephony/IVR systems and call processing; (ii) speech recognition and text-to-speech (TTS) engines and grammar authoring; (iii) web technologies (HTTP, HTML, CGI, Perl, browser client/server architectures); and (iv) relational databases. Such a person would be familiar with commercial speech platforms (Nuance, Lernout & Hauspie), voice-markup proposals (VoxML, VXML), and the then-common practice of "screen scraping" HTML with regular expressions in Perl/CGI. The '190 patent's own specification confirms this skill set: its preferred embodiment is built from a Nuance speech recognizer, an L&H TTS engine, dual-Pentium servers, and Perl-based agents (web_dispatch.pl, webget.pl) that use regular expressions to extract content.

3. Prior-art universe — and which listed references are not usable

The Google Patents page for the '190 patent carries a 140-entry "Citations" list. Only references with an effective prior-art date before February 4, 2000 qualify as § 102 prior art. Two entries in the Google Patents listing must be excluded:

• US6904450B1 (Geodata, "Method and system for customizable network data retrieval") — filed August 9, 2000, after the critical date; it appears in the "Families Citing this family" (later-citing) section, not the examiner's prior-art citations. Not prior art.
• US6937986B2 (Comverse, "Automatic dynamic speech recognition vocabulary…") — priority December 28, 2000; also in the later-citing section. Not prior art.
• US6434529B1 (Sun, "referencing object instances … from within a speech recognition grammar") — priority February 16, 2000, after the critical date. Not prior art (although it sits in the cited list, it cannot be used against the '190).

The usable prior art from the page's citation list includes, most importantly:

All of these are in the same fields of endeavor — telephony/IVR, speech recognition/synthesis, and web information retrieval — so they are analogous art under § 103.

4. The claim elements that matter

The core inventive footprint of the independent claims, distilled:

1. Setup phase: a user provides a URL (b); designates a pre-defined portion of that web page (c); identifies a named object associated with the content (d); the system generates a regular expression (a search-pattern text string) for the content (e); the user provides a speech command that is converted to digital form and assigned to the regular expression (f–h).
2. Retrieval phase: later, the user utters the same audio command (i); the system converts it (j), retrieves the associated regular expression (l), retrieves the live information from the pre-defined portion (m), and — if the information is not found there — searches the web site for the named object (n); the result is converted to audio and forwarded (o–q).

Claims 6 and 16 mirror this with system/instruction-set language: a server with telephony hardware, a speech-recognition engine, a speech-synthesis engine, a call-processing system, a first instruction set (URL + named object), a second instruction set that generates the regular expression, a recognition grammar, and a web browser comprising a content extraction agent, a content fetcher, and a content descriptor file.

5. Proposed combinations

Combination 1 (strongest): Motorola '336 + Talk2 '132 + HP '220

• Motorola '336 supplies the entire telephony/voice-browser infrastructure: server with telephony hardware, speech recognition (VRU server 234, STT 256), speech synthesis, call processing, grammars stored in a database (database server unit 244), barge-in, and spoken access to "designated city weather, traffic conditions, stock quotes, news updates" identified by URL. This maps to claim 1(a), (f)–(k), (o)–(q); claim 6's server/telephony hardware/recognition engine/synthesis engine/call-processing system/grammar/database; and claim 16's server/recognition grammar steps.
• Talk2 '132 supplies the missing "existing HTML, no re-authoring" piece that Motorola's VoxML-centric approach lacks. Talk2 explicitly diagnoses the problem the '190 solves: VoxML-only content must be authored separately, and "static translations" of HTML must be manually maintained whenever the source changes (Talk2, FIGS. 1–2 and Background). Talk2's solution — retrieve by URL, parse the document, identify categories (Weather, Portfolios, Headlines), filter out banners, and identify content by HTML tags/attributes — maps directly to claim 1(c)–(e), claim 6's first instruction set (URL + named object) and content descriptor file, and claim 21's named-object list (weather, forecast, high, low, radar, temp, humidity, wind, pressure, sunrise, sunset, time, month, day, stock, stock quote, news, news reel, airline, carrier, flight, flight number are all content categories of the kind Talk2 maps).
• HP '220 supplies the mechanism and motivation for binding a speech command to web-derived content: it builds a dynamic recognition vocabulary ("web-triggered word set") from the HTML source of the page being viewed, precisely because a small, content-specific vocabulary dramatically improves recognition accuracy — the same rationale the '190's Background gives ("quite accurate when a single, known user only needs to use a small vocabulary"). This maps to claim 1(f)–(h) (speech command → recognition grammar tied to content) and claim 16's assignment of a speech command to a recognition grammar corresponding to the instruction sets.

Why combine: The '190 solves a problem the prior art collectively identified but did not integrate: (i) phone access to the web was known (Motorola, Netphonic), (ii) voice access to unmodified HTML with automatic category/portion identification was known (Talk2), and (iii) dynamic, content-derived recognition vocabularies were known (HP). A PHOSITA building a "personal voice browser" in late 1999/early 2000 would naturally take Motorola's proven IVR/voice-browser platform, graft on Talk2's HTML-parsing/content-mapping engine so that users could reach existing pages (the very problem Talk2 was created to solve), and use HP's dynamic word-set technique to make the spoken commands accurate. That is a textbook KSR combination of known elements, each performing its known function, with a predictable result.

Combination 2: Netphonic '392 + Talk2 '132 + IBM '1472 / Motorola '266

• Netphonic '392 supplies telephone access to the Internet: a server fetches an HTML page by URL over HTTP, converts it to an H-tree, and reads the text aloud to the phone, with voice/DTMF navigation of links.
• Talk2 '132 again supplies content portioning (category mapping, tag/attribute-based identification, banner filtering).
• IBM '1472 (voice-command network navigation) and/or Motorola '266 (audio interface for document navigation) supply the speech-command recognition layer.

This combination is somewhat weaker than Combination 1 because Netphonic reads whole documents rather than user-selected portions, but when combined with Talk2's portion/category selection the combination still reaches the pre-defined-portion limitation.

Combination 3: Bell Atlantic '262 + Talk2 '132 + HP '220

• Bell Atlantic '262 supplies the audio access/conversion server for computer networks; Talk2 '132 supplies content portioning and tag-based identification; HP '220 supplies the content-derived recognition vocabulary. This combination emphasizes the "audio conversion" side of claims 1(o)–(q), 6, and 16.

The regular-expression limitation — an implementation detail

Claim 1(e)/claim 6's "second instruction set" require a "regular expression … a text string used for describing a search pattern." Regular-expression extraction of data from HTML was ubiquitous in the Perl/CGI server-side scripting of the era — indeed, the '190's own Tables 1–3 are Perl code whose content descriptor is literally a Regular_expression= field matched against fetched HTML (see webget.pl: @values = $content =~ /$Param{Regular_expression}[0]/). Given a goal of extracting a designated portion of a page (Talk2), the PHOSITA's standard tool for "find this content pattern in this text" was a regex. Under KSR, choosing the well-known tool (a regex) to implement a known function (locating a designated portion of an HTML document) is obvious. No single reference need recite the word "regular expression" for this limitation to be obvious; it is the conventional implementation of Talk2's parsing step, and the '190's own dependent claim language defines it no more narrowly than "a text string used for describing a search pattern."

The fallback limitation — the closest point, and how the combination reaches it

Claim 1(n) and claim 16's "searching said pre-selected web site … when the requested information is not found" is the most distinctive feature. The combination argument:

• Talk2 '132 already teaches identifying content by HTML tags/attributes rather than by fixed position — its "enhanced mapping" for My Yahoo! "looks for a tag with a particular background color attribute" to locate a category even as page layouts change. The '190's own specification describes exactly this rationale: named objects "may be moved within a document, but it still exists. Therefore, the content extraction agent 600 simply searches for the relevant name of the desired object."
• A PHOSITA implementing a content-extraction agent that uses a fixed location (the regex-derived pre-defined portion) would, as a matter of routine error recovery, add a fallback that re-parses the page for the named object when the primary extraction fails — particularly because Talk2 already teaches that tag/attribute-based identification is the robust method. This is "obvious to try" with a finite number of identified, predictable solutions (search for the named tag/attribute), which KSR treats as obvious.

I acknowledge this is the limitation a patent owner would defend hardest, and it is the one most likely to require the Talk2 '132 + general-knowledge rationale rather than an express teaching in a single reference.

6. Element-by-element mapping (claim 1)

Claim 6 maps the same content onto the system elements: Motorola '336 for the server/telephony hardware/recognition engine/synthesis engine/call-processing system/grammar/database; Talk2 '132 for the first instruction set (URL + named object), content descriptor file, and the parse/map/extract functions of the "content extraction agent"; the conventional webget.pl-style fetcher (the '190's Table 3) for the "content fetcher"; and HP '220 for the recognition-grammar/content binding. Claim 16 is a method-form restatement of claims 1 and 6 and is obvious for the same reasons.

Dependent claims: 2/20 (periodic updating) — Motorola '336 provides "up-to-date information" on each call; the '190 retrieves live each time. 3/5/17/19 (user-performed steps) — Talk2 '132's personal-home-page scenario; US20010032234A1 (user mapping of a document). 4/15/18 (graphical display and selection) — browser-based selection of page content was conventional (Talk2's category filtering presupposes a visual model of the page; the "clipping" UI is a standard browser-selection technique). 7–9 (landline/wireless/IP phone) — Motorola '336 and Netphonic '392 expressly contemplate wireline and cellular telephony; IP telephony was known by 2000. 10–12 (LAN/WAN/Internet) — Motorola '336 (electronic network 104 includes LANs/intranets/Internet). 13 (database) — Motorola '336 (database server unit 244). 14 (software to create instruction set from user-defined info) — Talk2 '132 (user-defined personal home pages mapped to categories); US20010032234A1. 21 (named-object list) — Talk2 '132's categories (weather, stock, news, etc.) are the same content labels.

7. Motivation and reasonable expectation of success

The KSR inquiry asks whether a PHOSITA would have had a reason to combine with a reasonable expectation of success. Here the motivations are explicit in the references themselves:

1. The problem is stated in the prior art. Talk2 '132's Background describes exactly the gap the '190 fills: phone users cannot conveniently access existing HTML; VoxML requires separate authoring; static translations require manual re-syncing as the source changes. The natural solution — dynamically parse existing HTML and present its relevant portions by voice — is Talk2's own contribution. Combining Talk2's parser with an existing voice-browser platform (Motorola '336) is the obvious engineering answer to the problem Talk2 identifies.
2. Recognition accuracy demands small, user-specific vocabularies. The '190's own Background concedes this. HP '220 teaches that the vocabulary should be derived from the web content itself and that doing so boosts accuracy. Binding a user-chosen spoken name to a clipped portion is the natural way to give each user a small, personal grammar — the precise teaching of HP '220 applied to Talk2's content categories.
3. Web pages change. Both Talk2 (tag/attribute-based category location) and the '190 (named-object fallback) address layout drift. A PHOSITA combining a fixed-location extractor with Talk2's tag-based identification would add the fallback as routine robustness engineering.
4. Predictable combination of known elements. Every component (telephony/IVR server, ASR/TTS engines, grammar databases, HTML fetching, regex-based extraction) existed and was commercially available (Nuance, L&H, VoxML, Perl/CGI). The '190's own disclosure relies entirely on these known components; the "invention" is their integration, which KSR treats as the paradigm of obviousness absent unexpected results.

8. Weaknesses and countervailing evidence (be candid)

• The fallback limitation (claim 1(n)/16) is the strongest defense point. No single cited reference expressly teaches "try the pre-defined portion first; if not found, search the site for the named object." The case for that limitation rests on Talk2 '132's tag/attribute-based identification plus routine error-recovery design — a legitimate but weaker combination argument. A district court or the PTAB could find this limitation non-obvious, particularly under a strict reading requiring a specific teaching of fallback logic.
• The real-world signal from the IPR record cuts against a slam-dunk obviousness case. Google's IPR2022-01020 against the '190 was denied institution on the merits, and Apple's IPR2022-01542 was terminated by settlement before institution (March 6, 2023). A merits denial at institution means the PTAB concluded the petition — which would have pressed the strongest available combinations, likely including several of the references above — did not establish a reasonable likelihood of prevailing on any claim. That is not a final merits determination and is not binding on a court, but it is probative that the best obviousness case the petitioner could assemble did not clear even the institution hurdle. (Sources: Google Patents litigation flags; portal.unifiedpatents.com/ptab/case/IPR2022-01020 and /IPR2022-01542; McKool Smith press release via tmcnet.com, Aug. 15, 2024.)
• Settlement of the Apple IPR means no merits decision there, and the district-court cases (N.D. Cal. 3:22-cv-07830, stayed Feb. 7, 2023; D. Del. 1:23-cv-00190) had not produced a § 103 merits ruling as of the information available to me.
• Dependent-claim nuances: claim 21's specific named-object list is arguably a selection of examples, but Talk2's categories cover the same semantic ground; dependent claims 4/15/18 (graphical display + selection software) are the conventional "clipping" UI and should be obvious over Talk2 '132's category mapping in combination with any browser.

9. Conclusion

Under a KSR-flexible § 103 analysis, the independent claims 1, 6, and 16 of the '190 patent would most likely be obvious over Motorola '336 (voice browser) in view of Talk2 '132 (voice interface for unmodified HTML with tag/category-based content identification) and HP '220 (content-derived dynamic speech vocabularies) — with Netphonic '392, Bell Atlantic '262, IBM '1472, and Motorola '266 as equally suitable voice-browser/telephony base references, and US20010032234A1 and Talk2 '6327572 as supporting the user-mapping/voice-service context. The dependent claims add only conventional implementation details (phone types, networks, databases, browser selection UIs, periodic updates, example named objects). The principal genuine vulnerability in the combination case is claim 1(n)'s named-object fallback, which depends on Talk2's tag-attribute identification plus routine-error-recovery reasoning; that is the limitation most likely to carry the patent if any claim survives, and it is consistent with the PTAB's refusal to institute Google's IPR on the merits. References with post-February-2000 priority dates (US6904450, US6937986, US6434529) must not be used as prior art against this patent despite appearing in the Google Patents listing.

Key sources: full patent text of US7516190B2 (Google Patents, provided in the prompt); FreePatentsOnline/Google Patents records for US5953392A, US6269336B1, US6349132B1, US5819220A, US6904450B1; Unified Patents PTAB case pages for IPR2022-01020 and IPR2022-01542; McKool Smith press release (Aug. 15, 2024) via tmcnet.com.