AUDIOVISUAL THREE-PASS REVIEW WORKSHEET Status: blank local review template Author disclosure: Original work by Alfred Purpose: Review one exact video candidate through picture-only, audio-only, and combined playback without turning one ordinary playthrough into a broad quality, rights, accessibility, platform, or publication claim. USE BOUNDARY This worksheet can record direct review of one identified local master. It does not by itself prove caption conformance, accessibility conformance, rights or policy clearance, deterministic reproduction, destination processing, public availability, audience response, customer activity, or revenue. Use NOT TESTED when a lane was not completed. Never convert missing evidence into PASS. Allowed lane results: PASS / FAIL / BLOCKED / NOT TESTED / SUPERSEDED Allowed overall decisions: HOLD / REVISE / LOCAL PLAYBACK REVIEWED / SUPERSEDED A blank field is incomplete evidence, not a pass. ====================================================================== A. REVIEW ENVELOPE ====================================================================== Review ID: Review revision: Reviewer role: Reviewer disclosure: Alfred Review started at, including timezone: Review completed at, including timezone: Decision this review may support: Decision this review may not support: Intended presentation class: Intended aspect ratio: Realistic review display size: Normal playback speed: Audio output path: Caption presentation used, if any: Required passes: [ ] Complete picture-only pass at normal speed [ ] Complete audio-only pass at normal speed [ ] Complete combined pass at normal speed [ ] Failure-boundary inspection Stop conditions: [ ] Candidate identity differs from the approved review input [ ] Playback cannot reach the complete final frame or final sound [ ] Required script, caption, or rights companion is missing [ ] Personal information, account data, contact details, secret, or credential appears or is heard [ ] Unreviewed third-party material appears or is heard [ ] A correction changes evidence needed by a completed pass [ ] The reviewer cannot distinguish an observed defect from a player defect ====================================================================== B. EXACT CANDIDATE IDENTITY ====================================================================== Stable candidate ID: Master filename: Master byte size: Digest algorithm: Complete master digest: Duration: Width x height: Frame rate: Container: Video codec: Audio codec: Audio channel layout: Approved script identity: Caption file identity: Render specification identity: Rights/provenance record identity: Claim-review record identity: Disclosure text: Identity checks: [ ] Digest was computed from the file opened for review [ ] Complete digest matches the approved candidate record [ ] Probe output describes the same file [ ] Script identity matches the candidate [ ] Caption identity matches the candidate [ ] Rights/provenance identity matches the candidate [ ] Disclosure is present in the intended surface Identity result: Evidence location: Finding or `none`: If identity does not pass, stop. Do not review convenient bytes and attach the result to a different candidate. ====================================================================== C. PLAYER AND OBSERVATION SETUP ====================================================================== Player and version: Local or destination-processed rendition: Display orientation: Display scaling: Audio output device class: Captions off/on/authored/generated: Overlays or masks intentionally shown: Unexpected controls or overlays: Setup limitations: [ ] Player can show the complete first frame [ ] Player can reach the complete final frame [ ] Mute state can be independently confirmed [ ] Picture can be hidden or attention removed for audio-only review [ ] Combined playback uses the intended audio route [ ] Review does not rely on a sampled contact sheet [ ] Review does not rely on a separate narration file [ ] Review does not claim to represent an untested destination rendition Setup result: Evidence location: Finding or `none`: ====================================================================== D. PASS 1 — COMPLETE PICTURE WITHOUT SOUND ====================================================================== Mute confirmation method: Pass start time: Pass end time: Reached first frame: YES / NO Reached final frame: YES / NO Normal speed maintained: YES / NO Review the complete timeline before diagnosing individual moments. [ ] Opening frame gives sufficient visual context [ ] Essential text is legible at the declared display size [ ] Essential text remains visible long enough to read [ ] Hook, claim qualifiers, and disclosure survive the intended frame [ ] Crops preserve essential labels, controls, and content [ ] Motion is intentional rather than a render jump or stale frame [ ] Cuts contain no blank, duplicate, or out-of-order frame [ ] Visual emphasis does not broaden or contradict the claim [ ] No personal information or account data is visible [ ] No contact detail, secret, credential, or private path is visible [ ] No unreviewed third-party media or copied interface is visible [ ] Final frame remains long enough to understand Timestamped picture findings: 1. Timestamp: Observed: Expected: Severity: Affected claim or surface: 2. Timestamp: Observed: Expected: Severity: Affected claim or surface: 3. Timestamp: Observed: Expected: Severity: Affected claim or surface: Picture-only result: Evidence location: Finding summary or `none`: Do not infer from this pass that audio, captions, rights, accessibility, or a processed destination rendition passes. ====================================================================== E. PASS 2 — COMPLETE AUDIO WITHOUT PICTURE ====================================================================== Picture-hidden or attention-removal method: Pass start time: Pass end time: Reached first sound: YES / NO Reached final sound: YES / NO Normal speed maintained: YES / NO Use the exact master's audio, not a separate narration or mix file. [ ] First syllable begins cleanly [ ] Spoken claims match the approved script [ ] Numbers, units, names, and technical terms are accurate [ ] Edits contain no click, repeated word, or unnatural gap [ ] Speech remains understandable under effects or music [ ] Relative levels do not change unexpectedly [ ] Meaningful non-speech sounds are clear and non-misleading [ ] Silence is intentional [ ] Audio does not depend on unidentified “this,” “here,” or “that” [ ] No personal information or account data is audible [ ] No contact detail, secret, credential, or private path is audible [ ] No unreviewed third-party audio is audible [ ] Final word and final sound are complete Timestamped audio findings: 1. Timestamp: Observed: Expected: Severity: Affected claim or surface: 2. Timestamp: Observed: Expected: Severity: Affected claim or surface: 3. Timestamp: Observed: Expected: Severity: Affected claim or surface: Audio-only result: Evidence location: Finding summary or `none`: Do not infer from this pass that picture, captions, rights, accessibility, or a processed destination rendition passes. ====================================================================== F. PASS 3 — COMPLETE PICTURE AND SOUND TOGETHER ====================================================================== Combined pass start time: Combined pass end time: Reached complete ending: YES / NO Normal speed maintained: YES / NO Caption implementation reviewed: [ ] Narration and the described visual appear in the intended order [ ] Effects correspond to the intended action and frame [ ] Speech and caption changes do not cross cuts misleadingly [ ] Visual labels agree with spoken numbers, units, and terms [ ] Music or effects do not obscure a key claim [ ] Repetition reinforces rather than changes meaning [ ] Pacing leaves time to understand text and action [ ] Disclosure is perceivable in the intended sequence [ ] Claim qualifiers retain their meaning in the combined presentation [ ] Captions do not collide with essential authored text or overlays [ ] Audio, picture, and captions finish coherently Timestamped combined findings: 1. Timestamp: Observed: Expected: Severity: Affected claim or surface: 2. Timestamp: Observed: Expected: Severity: Affected claim or surface: 3. Timestamp: Observed: Expected: Severity: Affected claim or surface: Combined result: Evidence location: Finding summary or `none`: Do not infer from this pass that caption syntax, flashing thresholds, codec conformance, rights, accessibility, platform policy, or public availability passes. ====================================================================== G. FAILURE-BOUNDARY INSPECTION ====================================================================== Complete normal-speed playback is the baseline. Focused inspection diagnoses short defects but does not replace the three complete passes. Record each applicable boundary: | ID | Timestamp/range | Boundary type | Expected | Observed | Result | |----|-----------------|---------------|----------|----------|--------| | 01 | | First frame | | | | | 02 | | First sound | | | | | 03 | | Cut | | | | | 04 | | Text change | | | | | 05 | | Fast motion | | | | | 06 | | Claim/number | | | | | 07 | | Caption cue | | | | | 08 | | Loud/quiet | | | | | 09 | | Disclosure | | | | | 10 | | Final frame | | | | | 11 | | Final caption | | | | | 12 | | Final sound | | | | Diagnostic method used: [ ] Normal-speed replay [ ] Frame stepping [ ] Slow playback [ ] Waveform or probe inspection [ ] Other: Diagnostic limitation: Boundary-inspection result: Evidence location: Finding summary or `none`: ====================================================================== H. CAPTION DEPENDENCY ====================================================================== Caption source: AUTHORED / DESTINATION GENERATED / NONE / NOT TESTED Caption identity: Caption language: Processed caption rendition available: YES / NO / NOT APPLICABLE [ ] Spoken words were compared with authored caption words [ ] Timing was checked against the exact master [ ] Line breaks and wrapping were reviewed [ ] Clipping and contrast were reviewed in the intended presentation [ ] Collisions with authored text and overlays were reviewed [ ] Speaker identity and meaningful sounds were handled as intended [ ] Generated destination captions remain NOT TESTED if unavailable Caption result: Evidence location: Finding summary or `none`: Caption review is separate evidence. A clean audio-only pass does not prove a caption file or destination caption rendition is correct. ====================================================================== I. FINDING AND CORRECTION LEDGER ====================================================================== Finding ID: Candidate digest: Pass and timestamp: Observed: Expected: Severity: Affected surfaces: Proposed correction: Correction owner role: Correction status: New candidate digest after correction: Evidence invalidated: [ ] Picture-only pass [ ] Audio-only pass [ ] Combined pass [ ] Boundary inspection [ ] Caption review [ ] Claim review [ ] Rights/provenance review [ ] Upload-selection evidence Required reruns: Rerun evidence: Finding terminal state: Repeat this block for every finding. Never edit an earlier pass so it appears to have reviewed changed bytes. ====================================================================== J. FAILURE-SHAPED METHOD TESTS ====================================================================== These tests exercise the worksheet, not a real release. 1. Review a file whose complete digest differs from the approved candidate. Expected: identity result FAIL or BLOCKED; playback approval is unavailable. Actual: 2. Complete one ordinary combined play and leave isolated passes blank. Expected: overall decision HOLD; blank passes do not inherit PASS. Actual: 3. Review a separate narration file instead of the master's audio. Expected: audio-only result BLOCKED. Actual: 4. Frame-step every cut but skip normal-speed playback. Expected: complete playback remains NOT TESTED. Actual: 5. Correct a visual title after all passes and retain the old approval. Expected: affected picture and combined evidence becomes SUPERSEDED. Actual: 6. Hear clean speech and mark captions PASS without reading them. Expected: caption result remains NOT TESTED. Actual: 7. Pass the local master and claim the destination transcode passed. Expected: destination rendition remains NOT TESTED. Actual: 8. Find identifying information in one transient frame. Expected: picture result FAIL and overall decision HOLD or REVISE. Actual: 9. Find unreviewed third-party audio under accurate narration. Expected: audio result FAIL; presentation quality cannot override rights. Actual: 10. Complete every pass but omit the complete master digest. Expected: overall decision BLOCKED. Actual: Method-test result: Finding summary or `none`: ====================================================================== K. FINAL DECISION ====================================================================== Picture-only result: Audio-only result: Combined result: Boundary-inspection result: Caption result: Identity result: Open findings: Blocked lanes: Not-tested lanes: Superseded evidence: Overall decision: Decision applies to complete master digest: Decision time, including timezone: Decision evidence location: Narrow supported statement, if all required lanes pass: “The identified local master received complete picture-only, audio-only, and combined normal-speed playback passes, plus focused checks at the recorded failure boundaries. Findings and corrections are attached to that exact candidate.” Do not use that statement if identity is incomplete, a required pass is blank, or changed bytes superseded the evidence. Explicitly not established: [ ] Caption conformance beyond directly recorded checks [ ] Accessibility conformance [ ] Rights or license clearance [ ] Platform policy clearance [ ] Deterministic reproduction [ ] Destination processing quality [ ] Upload acceptance [ ] Public visibility or logged-out availability [ ] Audience response or unique viewers [ ] Inquiry, customer, payment, or revenue result FINAL GATE [ ] One exact master and complete digest are recorded [ ] Script, caption, render, rights, claim, and disclosure identities agree [ ] Review setup and limitations are recorded [ ] The complete picture-only pass reached the final frame [ ] The complete audio-only pass reached the final sound [ ] The complete combined pass reached the complete ending [ ] Every completed pass ran at normal speed [ ] Focused diagnostics did not substitute for complete playback [ ] Findings use timestamps, observations, and expected results [ ] Privacy and identifying-data checks cover picture and sound [ ] Unreviewed third-party picture or audio produces a hold [ ] Caption evidence remains separate [ ] Local evidence is not transferred to an untested destination rendition [ ] Corrections supersede every affected pass [ ] Blank and unavailable evidence remain NOT TESTED or BLOCKED [ ] The decision names one exact candidate [ ] The supported statement stays inside the observed lanes [ ] No upload, publication, audience, customer, or revenue result is invented Final gate result: Reviewer role: Review completion time, including timezone: PROVENANCE NOTE This blank worksheet and its ten failure-shaped method tests are original work by Alfred. It contains no real customer material, account surface, destination result, upload, public video, audience response, customer activity, payment, or revenue evidence. A completed copy inherits the truth, rights, privacy, authority, security, retention, and publication duties of the material entered into it.