US 8755921

I'll search for the specific patent and any related litigation records.

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

I have solid patent data. Let me check for any litigation or CAFC involvement with this specific patent number before summarizing.

I have enough verified information from the searches (the litigation-specific search hit referenced a different patent number, which I will flag below, and no CAFC 2026 docket entry for this patent was found). Here is my summary.

US Patent 8,755,921 B2 — Summary

Patent number: US8755921B2 (application no. US 12/793,065)
Title: Continuous audio interaction with interruptive audio
Assignees: Google Inc. (original assignee, per the assignment recorded 2010-10-05); name changed to Google LLC (recorded 2017-10-02)
Inventors: Robert Hamilton (London, GB); Bjorn Bringert (Bath, GB)
Filing date: June 3, 2010
Issue date (grant/publication): June 17, 2014
Legal status (per Google Patents): Expired – Fee Related; adjusted expiration shown as 2031-02-24 (note: Google Patents labels this "an assumption and is not a legal conclusion")
Related filings: PCT/US2011/037728 (published as WO2011153025A1); continuation US 13/253,583 (published as US20120029672A1)
Claims/drawings: 39 claims; 6 drawing sheets

Abstract (verbatim): "According to an example implementation, a method may include playing, by a computing device, first audio generated by a first application, determining that the first audio generated by the first application should be interrupted based on a notification associated with a second application, pausing the first audio generated by the first application, playing interruptive audio generated by the second application during the pausing of the first audio, identifying a portion of the first audio that was previously played before the first audio was paused, and resuming the playing of the first audio so that the portion of the first audio that was previously played is replayed."

Plain-language overview of the claims

Claim 1 (independent, method) — full text available and confirmed. A first application executing on a computing device outputs continuous audio from a file. When it receives an interruptive notification from a second application, it pauses. The first application analyzes the audio to identify an earlier portion to re-output, and interruptive audio is output during the pause. On receiving a resume message, the first application automatically (with no user instruction) resumes at the location of the most recent tag in the file — a tag that is embedded in the file itself — re-outputting the identified earlier portion so the listener regains context.

In plain terms: when a phone call, navigation prompt, or alert interrupts a podcast/audiobook, the player doesn't just resume mid-thought; it rewinds to the last marked point in the file (e.g., a sentence or scene boundary stored in the file) and replays from there automatically.

Other independent claims — uncertainty noted. The issued patent contains 39 claims, but the complete text of the issued independent claims beyond claim 1 was not retrievable in my searches (the provided patent text and search results cut off at claim 2). Based on the published application (US20110301728A1) and the patent's own summary, the invention was drafted in parallel statutory categories, so the granted patent very likely includes independent claims for:

• an apparatus (processor + memory configured to perform the claim-1-like method);
• a non-transitory computer-readable storage medium that causes an application to pause on an OS interruption message, receive a resume message, and re-output the previously played portion; and
• a two-device method/apparatus/medium group in which a first computing device (e.g., a phone) receives a prompt, sends a pause message to a second device (e.g., a car stereo) that is playing continuous audio, outputs interruptive audio, then sends a resume message (optionally instructing a rewind/tag-based resume).

I could not verify the exact wording, numbering, or whether all of these survived prosecution as independent claims in the issued 39-claim set, so treat that breakdown as informed inference rather than confirmed text.

CAFC 2026 docket check

• No CAFC (or PTAB/district-court) docket entries for patent number 8,755,921 were found in my searches.
• One search returned Inventor Holdings LLC v. Google Inc. (D. Del., 1:14-CV-00186), but that docket's own filing refers to US 8,558,921 B2 — a different patent number. Per your instruction not to conflate similar numbers, I am not treating that case as related to 8,755,921.

Confidence notes / sources

• Bibliographic data is corroborated across Google Patents (patents.google.com/patent/US8755921), the USPTO-image PDF mirror, Justia, and uspto.report/patentsencyclopedia for the application publication.
• I could not confirm, from authoritative live sources, the full text of issued independent claims 2–39, any current maintenance-fee status beyond the Google Patents label, or any 2026 CAFC proceeding. If you need the complete issued claim set, the USPTO Patent Full-Text Image PDF (US8755921.pdf) is the authoritative source to pull next.

The initial searches returned nothing relevant. Let me query specialized patent-litigation databases directly.

Based on the searches I was able to run (against general web search and the Unified Patents litigation case list/analytics), I found no litigation involving US Patent 8,755,921 B2.

Result: No known litigation

For US Patent 8,755,921 B2 ("Continuous audio interaction with interruptive audio," assigned to Google LLC), I found no district-court complaints, no PTAB proceedings (IPR/PGR/CBM), and no CAFC appeals naming that patent number in any search I performed.

What I searched and found

Caveats and flags

1. Number-identification caution. The only superficially similar litigation I encountered anywhere was Inventor Holdings LLC v. Google Inc. (D. Del.), which involves US 8,558,921 B2 — a different patent number. Per your instruction not to conflate similar numbers, I am not treating that case as related to 8,755,921.
2. Search completeness. Several specialized docket sources (PACER, DocketNavigator, CourtListener full-text docket databases) could not be fully queried within the search limits I had. An absence of results from the sources I did reach is good evidence of no litigation, but it is not a definitive negative across every database.
3. Contextual plausibility. This is a Google-owned patent (inventors Robert Hamilton and Bjorn Bringert) that expired or went fee-related/expired around 2031 per Google Patents' adjusted-expiration label. Google-owned patents of this vintage are rarely asserted in litigation by Google itself, and I found no NPE assertion of this specific patent either.

Bottom line

As of April 26, 2026, based on the searches available to me, there is no known litigation involving US Patent 8,755,921. If you need a definitive negative, the next step would be a paid PACER/USPTO assignment-and-litigation search (e.g., DocketNavigator, LexisNexis PatentAdvisor, or RPX) keyed to the exact number 8,755,921 — and I'd recommend running a parallel check on the parent application publication US20110301728A1 and continuation US 13/253,583 (US20120029672A1) in case litigation was pled against the application family rather than the granted patent number.

Proceedings overview

There are zero AIA trial proceedings (IPR / PGR / CBM) on file for US Patent 8,755,921 B2: the USPTO Open Data Portal's structured listing returns no proceedings, and independent web searches for the patent number and title across PTAB decision databases surfaced none. For a defendant, the posture is a clean slate: no claims of 8,755,921 have been canceled or even challenged in any PTAB proceeding, so an IPR-based invalidity defense has not yet been tested — but the absence of any petition is also a signal that the patent has not been meaningfully asserted against parties willing to pay for a challenge.

No proceedings to detail individually — see the analysis below.

(No proceedings found)

• Type: N/A
• Filed: N/A
• Status: No AIA trial proceeding exists for this patent in the USPTO ODP ingest or in public search results (queries run 2026-09-05: "8,755,921" OR "8755921" inter partes review PTAB, "8755921" IPR petition patent trial, "US8755921" PTAB IPR final written decision, "Continuous audio interaction with interruptive audio" inter partes review, "8755921" site:uspto.gov OR site:unifiedpatents.com PTAB — all returned no relevant results; the only superficially similar hits involved unrelated patent numbers such as 8,927,592 / 8,558,921, which I am not treating as related per the number-identification rules).
• Judge panel: N/A
• Petition grounds: N/A
• Institution decision: N/A
• Final Written Decision: None issued — no claim of 8,755,921 has been through a PTAB trial.
• Settlement / termination: N/A
• Appeal: N/A — no FWD exists to appeal.
• Defensive value: There is no PTAB record either way. This means (a) no claims are dead — every claim in the 39-claim set stands as issued, and (b) no IPR has established a winning prior-art theory you can ride on. A defendant facing assertion cannot cite any FWD, but equally faces no estoppel from prior proceedings.

Strategic summary

Claim status — all UNTESTED. All 39 claims of 8,755,921 B2 remain as issued. Independent claim 1 (the only independent claim for which full issued text is confirmed in the record) requires: a first application outputting continuous audio from a file; pausing on an interruptive notification from a second application; the first application analyzing the audio to identify a previously-output portion to re-output; outputting interruptive audio during the pause; and, on a resume message, automatically re-outputting that identified portion and resuming at the most recent in-file tag (a tag embedded in the file itself). No PTAB petition has challenged this claim or any dependent claim, so there is no Board determination on how the "analyzing," "most recent tag," or "without user interaction" limitations will be construed, and no prior-art ground has been tested against them.

Estoppel landscape — wide open. Because no IPR/PGR/CBM has ever been filed on this patent, there are no § 315(e)(2) estoppel bars against anyone. Any defendant facing assertion today is free to raise any § 102/§ 103/§ 112 ground, on any prior art, in district court — nothing has been foreclosed. If a defendant chooses to file an IPR, the one-year statutory deadline from institution (35 U.S.C. § 316(a)(11)) and the § 315(b) one-year-from-service bar are the only timing constraints that will apply.

Pattern signals — none. There is no repeat petitioner, no patent-owner PTAB litigation pattern, no defensive-aggregator chain (e.g., Unified Patents) attached to this patent. Contextually, this is consistent with the patent's history: a Google-originated patent (inventors Robert Hamilton and Bjorn Bringert; filed 2010-06-03, granted 2014-06-17, assigned to Google LLC after the 2017 Google Inc. name change), showing a Google Patents "Expired – Fee Related" status label with an adjusted expiration of 2031-02-24. Google has not historically asserted this patent, and no NPE appears to have acquired and asserted it either — which is the most likely reason no petitioner has bothered to challenge it.

Recommended next steps

1. No PTAB lever to pull — do not cite one. If you are a defendant and any demand letter cites 8,755,921, there is no FWD to quote and no canceled claim to point to. Do not imply otherwise; the PTAB record is empty.

2. Confirm the enforcement posture before spending on an IPR. The absence of any IPR strongly suggests this patent is not being actively asserted. Verify (a) who currently owns it (assignment records show Google Inc. → Google LLC), (b) whether maintenance fees were paid or the patent is in fact expired/forfeit (Google Patents' "Expired – Fee Related" label and 2031-02-24 adjusted expiration are assumptions, not legal conclusions — check the USPTO Patent Center maintenance-fee window directly), and (c) whether the demand actually identifies this patent or merely a related family member (publication US20110301728A1; continuation US 13/253,583 / US20120029672A1; PCT WO2011153025A1).

3. If the patent is live and asserted, the IPR runway is open. No estoppel attaches, so a petition can present the strongest available § 102/§ 103 combination without concern for previously-raised grounds. Act quickly if the defendant was served more than a few months ago — § 315(b) bars a petition filed more than one year after service of a complaint alleging infringement of this patent.

4. Track the docket for future filings. Re-check the USPTO PTAB E2E and Unified Patents caselist periodically; an empty record today can change the day an NPE files suit. If a petition does surface, the statutory milestones to watch are: institution decision (due within 3 months of the petition's filing/preliminary response), and FWD (due within 12 months of institution, extendable by up to 6 months for good cause).

Confidence note: The "no proceedings" finding rests on the canonical USPTO ODP block in this prompt (which reports no AIA trial proceedings) corroborated by multiple web searches returning no relevant results. PTAB E2E full-text docket coverage and paid databases (PACER, DocketNavigator, RPX) were not exhaustively queried; if a proceeding exists that is not yet indexed in ODP or general web search, it would be a very recent filing, and re-checking USPTO PTAB E2E directly for the exact number 8,755,921 is the definitive next verification step.

I'll research the USPTO assignment records for US 8755921, including reel/frame numbers and correspondents.

Initial searches returned little. Let me try the USPTO assignment database directly and third-party indexes.

I need the reel/frame details. Let me try more targeted searches against assignment aggregator sites.

I've reached the search-step limit for this task, so I'll answer from the verified record data available (Google Patents legal-events feed, which mirrors USPTO assignment recordation for this patent) and flag precisely what could not be verified live.

Inventors

Pattern check: No unusual pattern is present. Both inventors executed the standard assignment to their employer, Google Inc., contemporaneous with filing (recorded 2010-10-05). I found no evidence that either departed Google shortly after filing; Bringert in particular appears in later Google patent records.

Original assignee

• Entity on the issued patent: Google Inc. (Mountain View, CA) — shown as original assignee on the 2014 grant; current assignee of record is Google LLC after the 2017 change of name.
• Primary line of business: Internet search, advertising, the Android mobile OS, cloud services, and consumer applications.
• Product embodiment: The claimed behavior — pausing a background continuous-audio app on an interruptive notification (call, navigation prompt, alert) and automatically resuming at a tagged context point — maps directly onto Android's audio-focus interruption handling as used by Google's own audio/ podcast/music apps. I cannot cite a specific shipped product version with high confidence from the searches available in this session; treat product embodiment as plausible but unverified by an explicit product citation.
• Current status: Operating — Google Inc. was renamed Google LLC effective 2017-10-02 (assignment recorded same day), and remains an operating subsidiary of Alphabet Inc. No bankruptcy, dissolution, or acquisition.

Assignment timeline

I could not retrieve reel/frame numbers or correspondent names from the live USPTO Assignment Center within my search limits (the Assignment Center is not fully indexable via the search tools available to me). The two events below are confirmed by the Google Patents legal-events feed for US 8755921, which is populated from USPTO assignment recordation. No third-party or post-issuance transfer appears in that feed.

• Executed ~2010-06 (contemporaneous with 2010-06-03 filing) / recorded 2010-10-05 — Reel not verified / frame not verified

  • Conveyance: ASSIGNMENT OF ASSIGNORS INTEREST
  • Assignor: ROBERT HAMILTON; BJORN BRINGERT (inventors)
  • Assignee: GOOGLE INC.
  • Correspondent: not retrievable in my searches (standard Google prosecution counsel expected; unverified)
  • Context: Ordinary inventor-to-employer assignment executed at filing — not an NPE-related transfer.

• Recorded 2017-10-02 — Reel not verified / frame not verified

  • Conveyance: CHANGE OF NAME
  • Assignor: GOOGLE INC.
  • Assignee: GOOGLE LLC
  • Correspondent: not retrievable in my searches
  • Context: Pure corporate renaming (Google Inc. → Google LLC) following the Alphabet Inc. reorganization; no change in beneficial ownership.

Finding: The recordation history visible for this patent consists solely of (1) the original inventors-to-Google assignment and (2) Google's own change of name. There is no recorded transfer to any LLC, holding company, licensing entity, or third party of any kind. Consistent with the prior litigation summary (no suits found naming this patent), the original operating assignee — now Google LLC — still owns the patent on the available record.

Timeline diagram

timeline
    title Ownership of US 8755921
    2010 : Filed by inventors
         : Assigned to Google Inc
    2014 : Patent granted
    2017 : Name change to Google LLC

NPE / troll-pattern signals

1. Shell-entity transfer — not present. The only "LLC" in the chain is Google LLC itself, the renamed operating company (change of name recorded 2017-10-02). No transfer to a licensing-only LLC, IP Holdings entity, or registered-agent address appears on the record.

2. Known asserter in the chain — not present. The chain contains only Google Inc./Google LLC. No Acacia, Marathon, Intellectual Ventures, Conversant, Vringo, or any Unified Patents/RPX-listed high-frequency plaintiff appears at any point.

3. Repeat correspondent across the chain — unclear. Correspondent names for both recordations could not be retrieved within my search limits. No recurrence finding is possible on the available data. (Both filings are routine Google corporate/in-house counsel filings; a single correspondent would not be probative regardless.)

4. Cascading transfers — not present. Exactly two recordations, seven years apart, both internal to Google. No chained-LLC sequence in a 24-month window.

5. Pre-litigation transfer — not present. No infringement suit naming this patent was found (see prior litigation summary), and no assignment is dated near any litigation event.

6. Bankruptcy fire-sale — not present. Google Inc./Google LLC has never been in Chapter 7 or 11; the 2017 recordation is a name change, not a distressed-asset sale.

7. Privateering — not present. No operating-company-to-NPE transfer exists on the record, so there is nothing to support a Google-sponsors-an-NPE theory.

8. Defensive aggregator (anti-NPE) — not present. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. It terminates at the original operating assignee (Google LLC). Note the inverse implication: a defensive aggregation was never needed because the patent never left Google's hands.

Verdict

Insufficient data

The only recordation events identifiable for US 8,755,921 are the original inventors-to-Google assignment (recorded 2010-10-05) and Google Inc.'s change of name to Google LLC (recorded 2017-10-02) — every available signal affirmatively indicates that Google LLC, the original operating assignee, still owns the patent and that no NPE/troll pattern of any kind is present. I nonetheless select "Insufficient data" rather than a stronger affirmative category because I could not verify reel/frame numbers and correspondent identities against the live USPTO Assignment Center within this session's search limits; that verification is the only missing piece before one could state with full confidence that the recordation history is complete. Verify at the USPTO Patent Assignment Search (https://assignmentcenter.uspto.gov/ — search by patent number 8755921) to confirm the two entries above and to pull reel/frame and correspondent data.

I was not able to retrieve the USPTO's "References Cited" (backward-citation) list for US 8,755,921 B2 within the search budget I had for this task. I need to be transparent about that before giving you an analysis, because an anticipatory-prior-art answer that named specific patent numbers from memory would risk fabrication — which I will not do. Here is exactly what I verified, what I could not verify, and an analytical framework you can use once the authoritative citation list is pulled.

What the searches actually returned

Verified bibliographic data (corroborated):

• US 8,755,921 B2 — "Continuous audio interaction with interruptive audio"; inventors Robert Hamilton and Bjorn Bringert; original assignee Google Inc. (now Google LLC); filed 2010-06-03; granted 2014-06-17; 39 claims.
• Application family: US 12/793,065 → published as US 2011/0301728 A1; continuation US 13/253,583 (US 2012/0029672 A1); PCT/US2011/037728 (WO 2011/153025 A1).

What I could not retrieve:

• The front-page "References Cited" list of the issued patent (the examiner-cited U.S. and foreign patent documents and non-patent literature). The Google Patents page text provided in the task does not include the Citations section, and my targeted searches for the citation list of 8755921 did not surface it. I also could not pull the PCT International Search Report for WO 2011/153025 A1, which would have contained the closest art cited against the family.

Forward citations verified in search results (later documents citing 8,755,921 — these are not prior art against it, but confirm its relevance and are listed so you don't confuse them with § 102 art):

• US 2021/0216272 A1 — "Audio Content Auditioning by Playback Device" — cites US8755921B2.
• US 2017/0235545 A1 — "Leaving Group And Smart Line-In Processing" — cites US8755921B2 (its citation list also shows US 2011/0316768 A1 to Vizio).
• JP 5740033 B1 — "Information processing device, portable terminal device, and program" — cites US8755921B2.
• Pages for US 2009/0299505 A1 ("Optimizing seek functionality in media content") and US 7,178,047 B2 show US8755921B2 in their citation lists (i.e., those patents/applications are among the later documents citing 8,755,921).

Why I won't guess the examiner-cited list

The task asks me to analyze "each patent citation for 8755921" and map each to specific claims under § 102. Doing that properly requires the actual citation list from the patent's first page (USPTO full-text image PDF) or Patent Center. I do not have that list confirmed from a live source, and the operating rules I'm under require that I flag low-confidence items rather than present inferences as fact. If I named prior-art patent numbers from training memory, I would risk auto-correcting or conflating identifiers — exactly what I'm instructed not to do.

What the claim analysis will need to map against (framework)

Once the citation list is retrieved, the § 102 analysis for this patent will hinge on claim 1 (method) and its siblings (apparatus claim ~11, medium claim ~21, streamed-content method claim ~31, per the Justia text surfaced above), which share these core limitations:

1. First application outputs continuous audio from a file it reads (audio and optionally video; claims 2, 22, 26).
2. Receives an interruptive notification (from an OS temporary-interruption message or a second application).
3. Pauses the continuous audio (optionally with fade-out/fade-in, claims 23; storing streamed data or messaging the sender to pause/rewind, claims 24, 31–33).
4. The first application analyzes the audio to identify a previously-output portion to re-output (content indexing / sentence-boundary detection / scene-change detection, claims 25–26).
5. Interruptive audio is output during the pause (phone, navigation, calendar/email/text alert — claims 28, and original dependent set).
6. On a resume message, the audio automatically re-outputs the identified portion without user interaction, starting at the most recent in-file tag (a tag embedded in the file), restoring context.

For each candidate reference, the § 102/103 question will be whether it discloses all of: (a) tag-based resume to a pre-pause point, (b) the tag being in-file/embedded, (c) automatic replay without user instruction, and (d) the application-side analysis (e.g., sentence boundary or content-index detection) used to select the replay point. The most likely art categories to look for in the citation list, and the claims they would threaten:

• Audio-ducking / phone-interrupts-music systems (likely relevant to claims on pausing continuous audio during a call/notification — i.e., the base method/apparatus/medium claims, but often lacking the replay-with-tag element).
• Bookmarking/rewind-on-resume players (TiVo-style "overshoot correction" and "bookmarking" references are the classic art here; relevant to the replay/resume limitations of claims 1, 11, 21, 31, and their dependents).
• Speech/sentence-boundary detection and content-indexing publications (relevant to the "analyzing … identify a portion" and "complete sentence … content indexing" dependents, e.g., claims 25, 30-type limitations).
• Turn-by-turn navigation and notification-priority systems (relevant to the second-application/notification limitations, claims 3–10-type).
• Two-device remote-control pause/resume schemes (relevant to the claim group drawn to a first device sending pause/resume messages to a second device — the FIG. 3 embodiment, claims ~34–39 area).

Recommended next step to get the authoritative list

The definitive source is the USPTO full-text image PDF of US 8,755,921 (page 1 contains the "(56) References Cited" block) or USPTO Patent Center for the application 12/793,065. The PCT search report for WO 2011/153025 A1 is a second authoritative source for the closest art. If you can paste that reference list (U.S. patents, foreign patents, and NPL), I can immediately produce the per-reference table you asked for — full citation, dates, description, and the specific claims each reference potentially anticipates under § 102 — using the claim map above.

Confidence note: High confidence on the bibliographic data and the claim structure of the independent claims (claims 1, 11, 21, and 31, plus selected dependents, were visible in the Justia text returned by search). Low/unknown confidence on the identity of the examiner-cited references for 8,755,921 — that data point was not retrievable in my searches, and I have deliberately not guessed it.

Important threshold note

The task directs me to use "the results from the Prior Art section of this page," but no Prior Art section was included in the material provided to me. The message contains only the Patent Summary, Litigation Summary, PTAB Challenges, Strategic Summary, and Recommended Next Steps. I could not verify the full set of references cited on the face of US 8,755,921 B2 (the Google Patents "Citations" block and the USPTO file wrapper were not retrievable within my search limits). The analysis below is therefore an analytical framework built on (a) the confirmed issued text of claim 1, (b) the specification, and (c) representative, well-documented pre-2010 references in this field — not a verified recitation of the examiner's actual § 103 rejections. Where I identify specific references by number, their relevance descriptions are grounded in what I know with reasonable confidence, and I flag anything that needs verification before it is relied on in a real pleading.

Claim 1 element breakdown (confirmed issued text)

Claim 1 requires, in order:

1. A first application executing on a computing device outputs continuous audio corresponding to a file.
2. The first application receives an interruptive notification associated with a second application (the notification corresponds to interrupting the continuous audio).
3. The first application pauses the continuous audio based on that notification.
4. The first application analyzes the continuous audio to identify a portion previously output, to be re-output.
5. Interruptive audio associated with the second application is output during the pause.
6. On receiving a resume message upon completion of the interruptive audio, the first application resumes at a location associated with a most recent tag in the file, and automatically re-outputs, without user interaction, the identified portion — with the tag being included in the file from which the audio is output.

Note that the phrase "automatically re-outputting, without user interaction or instruction" and the requirement that the tag be "included in the file" are prosecution-era narrowing features relative to the as-published application (US20110301728A1), whose claims required neither. That narrowing (embedded, in-file tags) is both the most likely reason the claim issued and the most defensible point of distinction — and it should anchor the obviousness analysis.

Person having ordinary skill in the art (PHOSITA)

A PHOSITA as of June 3, 2010 would be someone with 2–4 years of experience in mobile-OS audio management, multimedia player design, and audio coding (e.g., an engineer working on smartphone audio-focus/audio-interruption frameworks or embedded media players), with working knowledge of: telephony call handling, audio focus/ducking policies, streaming and buffering, media file container formats and metadata (ID3, chapter markers, cue sheets), and speech/audio analysis for boundary detection.

Representative prior-art landscape (pre-June 2010)

These are the art areas a § 103 challenge would draw on. Each maps to a claim-1 element:

A. Interruption/ducking/resume of background audio on phone call or notification. By 2010 it was commonplace in feature phones and early smartphones to duck or pause music/podcast playback when a call or alarm arrives and to resume afterward (Symbian, Windows Mobile, and early Android/iOS audio-focus behavior; numerous handset patents). This covers elements 2, 3, and 5 essentially directly.

B. Automatic rewind/replay of a preceding portion upon resume to preserve user context. This is the conceptual heart of element 6's re-output requirement. Representative, verifiable references:

• US 6,850,691 B1 ("Automatic playback overshoot correction system," TiVo) — teaches that after an interruption/pause, playback should resume from a point before the pause so the user does not miss content; i.e., automatic rewind of a preceding portion upon resume. (This reference surfaced in my searches in the citation block of related-family documents, and its title/subject matter are consistent with that role.)
• US 2007/0230921 A1 ("Multimedia time warping system," Barton/TiVo) — teaches pause, buffering, and resumption of a live media stream with time-shifted continuation, including re-presenting buffered content after interruption.
• US 6,868,225 B1 ("Multimedia program bookmarking system," TiVo) — teaches storing resume points/bookmarks so playback can continue from a stored location after an interruption.

C. Content markers/tags embedded in media files. By 2010, audiobook and podcast players routinely used embedded chapter/section markers (e.g., chapter tags in MP4/AAC audiobooks, ID3 chapter frames, podcast chapter marks, DVD chapter stops), and speech-analysis tools performed sentence-boundary detection, content indexing, and transcript alignment. This covers element 4 ("analyzing … to identify a portion") and the "most recent tag … included in the file" limitation. A PHOSITA would know of both (i) tags that ship inside the file (container-level markers) and (ii) on-the-fly detection of sentence/scene boundaries from audio energy, silence, speech recognition, or transcripts — all of which the specification itself lists (col./¶ discussing "tagged with sentence beginnings … based on pauses or audio energy, sentence boundary detection … content indexing … transcript").

Proposed combinations under § 103

Combination 1 — Interruption-aware media player + automatic-rewind-on-resume + embedded content tags

• Primary reference: a mobile device/OS teaching that an audio player pauses background continuous audio (music/podcast/audiobook) when a second application (phone, navigation, calendar) requests audio output, then resumes the paused audio after the interruptive audio completes (prior-art group A).
• Secondary reference: TiVo-style automatic playback-overshoot correction / time-warping resume (US 6,850,691; US 2007/0230921; US 6,868,225) teaching that on resume after an interruption, playback should start at a point preceding the interruption (a bookmark or a rewind offset) so the user regains context.
• Tertiary reference: an audiobook/podcast player or media-container reference teaching chapter/sentence markers embedded in the file, and/or speech-analysis references teaching sentence-boundary detection and content indexing.

Element-by-element mapping:

Motivation to combine (the § 103 rationale): The problem the patent identifies — "transferring from the continuous audio to the interruptive audio and back … may lose the context or continuity" (Background) — was a known, long-felt annoyance in mobile media playback. Each piece of the solution was a known, independently functioning component: pause-on-interrupt was standard OS audio policy; resume-with-overshoot-correction was expressly solved by TiVo for time-shifted viewing; and content markers/boundary detection were standard in audiobook/podcast tooling. A PHOSITA seeking to preserve listener context would have had an obvious reason to combine them: replace a blind "resume at the exact pause point" with "resume at the most recent content boundary," because that is the simplest way to give the listener a syntactically complete restart point. This is the kind of "combination of familiar elements according to known methods" that KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), treats as prima facie obvious when the result is the predictable one (restored context/continuity). The design choices — sentence boundary vs. fixed 5/10-second rewind vs. scene start — are all disclosed in the patent itself as alternatives, which itself evidences that a PHOSITA would treat them as interchangeable known options.

Combination 2 — "Duplication/redundancy" fallback: resume-point bookmarking with fixed rewind instead of analyzed tags

Even if the "analyzing" and "embedded tag" limitations were given weight, an alternative ground would combine group A (interrupt-aware pause/resume) with a fixed-offset rewind-on-resume reference (e.g., US 6,850,691's overshoot correction or a car-stereo/radio "rewind on pause/resume" feature). The dependent-claim set (claims 2–39, which I do not have full text for) may be more vulnerable still, since the specification's own dependent concepts — fixed 5/10-second rewind, gradual volume fade-in/fade-out ("ducking"), buffering streamed data during the pause, resuming a complete sentence determined by content indexing — are each individually well-trodden pre-2010 techniques. Any analysis of dependent claims should await the full issued claim text, which the previous sections flagged as unverified.

Where the obviousness case is weakest (defense-side counterpoints)

A patent owner would have credible responses, and they should be anticipated:

1. "Without user interaction or instruction" (automaticity). The re-output must occur automatically on the resume message, with no user prompt. If the TiVo-class references require user confirmation or involve user-initiated replay (e.g., a "replay" button), the challenger must show a reference where the replay is automatic on interruption-resume. US 6,850,691's entire premise is automatic correction of overshoot, which is the best candidate, but its context (user reaction time after a manual pause) differs from an application-initiated interruption — the challenger needs to articulate why the distinction is not meaningful.

2. "Most recent tag in the file … included in the file." This is the narrowest limitation. The tag must be a marker embedded in the file the first application reads — not merely a separately computed bookmark or an OS-level timestamp. To win, the challenger needs a reference disclosing (i) file-embedded chapter/sentence markers and (ii) resuming at the most recent such marker after an interruption. Audiobook chapter-marker players are the obvious source, but the examiner evidently did not find (or did not rely on) a single reference squarely disclosing interruption + automatic resume at the most recent in-file tag; that is presumably why the claim issued in this form.

3. No single reference teaches the full sequence. The claimed method is a specific ordered chain (pause → analyze → play interruptive audio → on resume message, jump to in-file tag → auto re-output). A § 103 case will therefore stand or fall on the strength of the motivation showing, not on any single anticipating reference. The challenger should anchor the motivation in the patent's own Background (context loss is the problem) and in industry literature from 2005–2010 on audio-focus policies (e.g., Nokia/Symbian and early Android audio-interruption design discussions), which treated "duck or pause, then restore" as the standard.

Bottom line and recommended verification steps

• Prima facie obviousness is plausible but not slam-dunk. The core concept — pause continuous audio for an interruptive application, then resume by replaying enough earlier content to restore context — is a predictable combination of (A) standard audio-focus interruption handling, (B) known resume-overshoot correction, and (C) known content-boundary markers/analysis. The two features most likely to carry the patent are automaticity (no user instruction) and the embedded, most-recent-tag resume point; a serious § 103 ground must aim squarely at those.
• Do not plead specific prior-art numbers without verification. Because the actual "Prior Art section" was not provided and I could not retrieve the face-of-patent citations or file wrapper, treat US 6,850,691, US 6,868,225, and US 2007/0230921 as candidate references whose full disclosures must be pulled (USPTO Patent Center / Google Patents) and checked for: automatic vs. user-initiated replay, file-embedded tags, and application-to-application interruption (as opposed to user pause). Any IPR or litigation ground should also be checked against the one-year § 315(b) clock noted in the earlier sections.
• Highest-value next step: pull the prosecution history (USPTO Patent Center, application 12/793,065) to see exactly what the examiner cited, what amendments produced the "automatically … without user interaction" and "tag … included in the file" language, and what arguments overcame the prior art. That record will show the examiner's own view of the closest art and the precise distinctions that got claim 1 allowed — which is the roadmap for the strongest § 103 combination and for rebutting it.

Confidence: High on the claim-1 text, the specification content, and the overall structure of the obviousness argument. Medium-to-low on the specific prior-art references (unverified details and applicability), and I could not assess claims 2–39 because full issued text was not confirmed. No representation is made that any named reference is the same art the examiner considered.