# Find transcript excerpts with checked page and line references Canonical page: https://agents.paretoid.com/workflows/find-transcript-excerpts-with-checked-page-and-line-references/ Category: Operations & Reporting Practice relevance: cross-practice. Useful supporting administrative method; no claim of practice-specific legal instructions. Jurisdiction/provider scope: Adobe Acrobat searchable transcripts; verify original page/line references, no transcript-provider integration. Organize supplied search hits without inventing transcript coordinates. Status: Proposed / not tested. The public source describes the problem. Documentation supports only the stated method components; no result from this recipe has been verified. Verification: Source methods and reported problems have been reviewed. We have not run the adapted workflow. The sample below describes expected behavior on synthetic data, not a customer result. ## The problem Organize supplied search hits without inventing transcript coordinates. Why it matters: A useful excerpt can become misleading if it loses the surrounding answer or points to the wrong page. Checking original transcript labels and line numbers helps the lawyer locate the passage while preserving OCR and context limits. ## Sources and evidence - pain: [Pain evidence: Billing: fair rule of thumb for reviewing transcripts?](https://www.reddit.com/r/LawFirm/comments/14oe7e1) Billing: fair rule of thumb for reviewing transcripts?, 2023-07-02, older: an of-counsel lawyer writing briefs for cases whose depositions they did not attend says working through transcripts takes a long time. Practitioner replies emphasize that search cannot replace reading the context. The direct page intermittently failed to open; the indexed original thread exposed the dated full post and comments. Reproduced on 2026-10-15 with the search query “Reddit lawfirm billing reviewing deposition transcripts briefs not present deposition July 2023”; indexed title: “Billing / What's a fair billing rule-of-thumb for reviewing transcripts?” The original dated post says: “it takes me a really long time to work through the transcripts. - method: [Method: Adobe Acrobat searching PDFs](https://helpx.adobe.com/acrobat/using/searching-pdfs.html) Adobe Acrobat searching PDFs, updated 2026-02-26, documents desktop advanced search across PDFs, exact-phrase queries, contextual results and CSV/PDF result export. It does not promise automatic certified transcript page/line mapping, which must be checked against the original. Source checked: October 15, 2026. Pain accounts are self-reported. Documentation supports components; this adapted recipe and prompt have not been tested. ## Suggested approach Start when: Attorney provides a bounded list of topics/terms and approved transcripts 1. search locally, collect exact excerpts with surrounding questions/answers and verify printed transcript coordinates 2. Keep source references and report unknowns instead of filling gaps. 3. The responsible reviewer checks the report before any separately approved action. ## What you need - Actual authorized transcript PDF or faithful typed transcript, source/version IDs, lawyer-approved topic list and a reviewer who can read original printed page/line numbers. - Existing firm-approved Acrobat Studio, AI Assistant add-on, or another plan including Acrobat AI; online services/cloud storage, language and region must be permitted. No purchase, login or connection is part of this verification. The manual route works without AI. - Preserve original evidence unchanged in restricted storage. Use synthetic material first. Prepare a separate, explicitly approved minimized AI input; aliases alone do not anonymize narrative or attachments. Do not upload raw originals or an alias key merely because another tab/file is sanitized. - Use synthetic inputs first. Keep real data in firm-approved accounts and processing, with appropriate access, confidentiality and retention controls. Aliases alone do not anonymize records. Never paste credentials, API keys or unnecessary client details. ## How to use Documentation-verified component path; proposed adaptation, not tested - AI citation numbers and PDF page indexes are not transcript page/line citations. A transcript without trustworthy original coordinates blocks final citation output. Access and setup: - Actual account access, source versions/completeness, privacy approval and reviewer sign-off were not observed. ### 1. Acquire and map the actual transcript Tool: Acrobat desktop Location: Open transcript > page navigation and Select text Preserve the actual transcript unchanged and record source/version, page count, draft/final/errata status and the approved search scope. Use an explicitly approved bounded transcript PDF or faithful typed transcript for the first AI search; do not manually identify all relevant pages first. A synthetic/public transcript is the pilot default. If privacy approval excludes a full transcript, staff must supply a bounded approved excerpt set and label that reduced coverage. Do not invent original page or line numbers. Expected output: Actual approved bounded transcript source, topic list and scope/version manifest; original coordinate map may be incomplete until candidate hits are checked. Next step: Load the approved transcript source and topics first; locate candidates, then verify original page/line coordinates for those hits locally. Check before continuing: Page-number collisions, rough drafts, errata and missing line numbers stay explicit; exact locator may be null. Execution status: Official public documentation checked 2026-10-15; no account, source-data acquisition, prompt run or end-to-end workflow was tested. Step documentation: https://helpx.adobe.com/acrobat/using/reusing-pdf-content.html https://helpx.adobe.com/acrobat/desktop/get-started/learn-the-basics/navigation.html ### 2. Load only the approved AI input Tool: Acrobat PDF Spaces Location: Acrobat homepage > PDF Spaces > Create a PDF Space > upload files or Paste copied text > Add to PDF Space Add the explicitly approved bounded transcript PDF or faithful typed transcript, topic terms and source/scope/version manifest. Include any already verified coordinate map, but do not fabricate one before candidate search. Use clear source/version labels and verify the Files panel. Do not add originals that have not separately been cleared for Adobe processing. Expected output: Visible approved source set in the PDF Space; no account integration is implied. Next step: Use the AI Assistant message field in this same space for the next prompt. Check before continuing: Confirm every named source is actually loaded and readable. PDF Spaces supports up to 100 files, 100 MB and 600 pages per file; unsupported/password-protected/handwritten material blocks affected extraction. Execution status: Official public documentation checked 2026-10-15; no account, source-data acquisition, prompt run or end-to-end workflow was tested. Step documentation: https://helpx.adobe.com/acrobat/desktop/explore-pdf-spaces/create-pdf-spaces.html https://helpx.adobe.com/acrobat/desktop/explore-pdf-spaces/supported-formats-limitations.html https://helpx.adobe.com/acrobat/desktop/explore-pdf-spaces/pdf-spaces-faq.html ### 3. Run the bounded prompt and inspect citations Tool: Acrobat AI Assistant Location: The PDF Space AI Assistant message field Paste the prompt below. Open each source-number citation, check it, and use Copy to transfer the answer into a new local Word document or text file. Correct errors using the originals before reliance. Prompt: ```text Locate candidate excerpts for the supplied topic list using only these transcript sources. Return topic_id,transcript_id,version,candidate quotation,source block,PDF page,original printed page,start_line,end_line and uncertainty. Use original page/line values only when supplied and verified; otherwise set them null. Include nearby question/answer context and distinguish speaker labels. Do not invent coordinates, re-number lines, summarize testimony as established fact, determine credibility or legal relevance, or change a transcript. A candidate hit requires human original-page verification. ``` Expected output: Candidate excerpt list with supported coordinates or nulls and context. Next step: Save the checked draft as transcript-excerpt-review.docx; carry source IDs, versions and reviewer corrections into the next review step. Check before continuing: Open the actual original page for every hit; verify literal quote, speaker, page, line span and any errata. Execution status: Official public documentation checked 2026-10-15; no account, source-data acquisition, prompt run or end-to-end workflow was tested. Step documentation: https://helpx.adobe.com/acrobat/desktop/use-acrobat-ai/generative-ai-features/ai-get-answers.html ### 4. Human review and final handoff Tool: Responsible staff reviewer and lawyer/process owner Location: Local reviewed report beside the frozen source set Reviewer manually searches the original using a distinctive phrase, reads the full question/answer, and writes verified original page:line spans into the local list. Mark unverified coordinates unresolved. Only now complete the coordinate map for verified hits, retaining PDF index separately from printed page and original line span. Unknown coordinates remain null. Expected output: Verified locator list for lawyer review, with coverage and version. Next step: Lawyer decides whether and how an excerpt may be used; no quote is inserted into a filing automatically. Check before continuing: All unresolved or unreadable evidence stays open. No inferred approval or external action follows from AI output. Execution status: Official public documentation checked 2026-10-15; no account, source-data acquisition, prompt run or end-to-end workflow was tested. Step documentation: https://helpx.adobe.com/acrobat/desktop/use-acrobat-ai/generative-ai-features/ai-get-answers.html ## What you will get excerpt-locator.csv: hit_id, topic_id, transcript_id, version, exact_excerpt, surrounding_question_answer, printed_page, line_start, line_end, pdf_page, coordinate_status, truncation_flag, ocr_flag, source_locator. No credibility or legal-relevance judgment. ## Human review and limits Attorney/paralegal verifies every excerpt against the original and reviews adequate context. Test a four-up PDF, an answer spanning pages, a negated phrase and OCR corruption. Distinct from S01 institutional knowledge retrieval and citation QA of already-written drafts. The assistant drafts or reports only. Check source records and review the output before a human performs any separately approved send or change. No automatic legal decision, money movement, account connection or publication is included. What to measure: In a synthetic pilot, count correctly found exceptions, false alarms, and unresolved inputs. No time or cost saving is assumed. ## Original prepared-input schema topics.csv: topic_id, approved_topic, approved_terms transcripts.csv: transcript_id, version, original_locator, layout, coordinate_review_owner search-hits.csv: hit_id, topic_id, transcript_id, version, exact_excerpt, surrounding_question_answer, printed_page, line_start, line_end, pdf_page, coordinate_status, truncation_flag, ocr_flag, source_locator; scope.txt: search method, search coverage and missing-hit limitations Synthetic example input: ```text Synthetic inputs are included in the starter kit. topics.example.csv transcripts.example.csv search-hits.example.csv scope.example.txt ``` Expected synthetic output: ```text H1: retain exact negated wording "I did not see a delivery." and printed page 12 lines 4-5; PDF page 3 stays a separate locator. Do not summarize as saw delivery. H2: retain partial excerpt and flag truncated answer plus unverified coordinates. Do not invent continuation or line range on next page. H3: keep exact co1or/b1ue OCR text; missing printed coordinates and OCR uncertainty stay visible. Do not silently repair wording. TOP2 has an OCR-corrupt supplied hit; do not conclude the topic is absent or coordinates verified. All excerpts remain attorney-review locators; no credibility or legal-relevance decision. ``` Failure check: Attorney/paralegal verifies every excerpt against the original and reviews adequate context. Test a four-up PDF, an answer spanning pages, a negated phrase and OCR corruption. ## Starter files Native blank scaffolds and original synthetic examples may use different schemas. Follow NATIVE-INPUT-MAPPING where supplied. These files do not connect accounts or install automation. - [Download starter kit](https://agents.paretoid.com/assets/starter-kits/s25-starter-kit.zip) — All inputs, synthetic examples and review notes in one ZIP. No automation is installed. - [Start here: get the sources and prepare the review](https://agents.paretoid.com/assets/starter-kits/s25/START-HERE.txt) — Documentation-checked acquisition, explicit manual mapping, exact prompts and human review limits - [topics.csv](https://agents.paretoid.com/assets/starter-kits/s25/topics.csv) — Blank input template with exact recipe headers. - [topics.example.csv](https://agents.paretoid.com/assets/starter-kits/s25/topics.example.csv) — Separate synthetic fixture input. Do not mix these rows into real exports. - [transcripts.csv](https://agents.paretoid.com/assets/starter-kits/s25/transcripts.csv) — Blank input template with exact recipe headers. - [transcripts.example.csv](https://agents.paretoid.com/assets/starter-kits/s25/transcripts.example.csv) — Separate synthetic fixture input. Do not mix these rows into real exports. - [search-hits.csv](https://agents.paretoid.com/assets/starter-kits/s25/search-hits.csv) — Blank input template with exact recipe headers. - [search-hits.example.csv](https://agents.paretoid.com/assets/starter-kits/s25/search-hits.example.csv) — Separate synthetic fixture input. Do not mix these rows into real exports. - [scope.txt](https://agents.paretoid.com/assets/starter-kits/s25/scope.txt) — Blank text input worksheet. Fill its fields with approved evidence; no default assumptions are supplied. - [scope.example.txt](https://agents.paretoid.com/assets/starter-kits/s25/scope.example.txt) — Separate synthetic text input fixture; use with the example CSV files only. - [Expected fixture findings](https://agents.paretoid.com/assets/starter-kits/s25/expected-fixture-result.txt) — Read-only expected output and failure conditions for the included synthetic input files. - [Setup and review checklist](https://agents.paretoid.com/assets/starter-kits/s25/README.txt) — Recipe-specific prerequisites, file mapping, prompt, setup steps, checklist and limits. ## Optional original prepared-input prompt Destination: Your firm-approved AI assistant · prepared files This original prompt uses its own prepared-input schema. Do not mix it with native-step tables without the supplied mapping. ```text Organize these approved transcript search hits under only the supplied topics. Preserve exact wording, transcript ID/version, printed page and line references, separate PDF page locator and surrounding question/answer context. Never invent missing lines or silently fix OCR. Flag every unverified coordinate and possible truncated answer. Do not assess witness credibility, decide legal relevance, generate testimony or suggest that unreturned hits prove absence. This is a locator for attorney review, not a substitute for reading the transcript. INPUTS topics.csv: topic_id, approved_topic, approved_terms transcripts.csv: transcript_id, version, original_locator, layout, coordinate_review_owner search-hits.csv: hit_id, topic_id, transcript_id, version, exact_excerpt, surrounding_question_answer, printed_page, line_start, line_end, pdf_page, coordinate_status, truncation_flag, ocr_flag, source_locator; scope.txt: search method, search coverage and missing-hit limitations OUTPUT excerpt-locator.csv: hit_id, topic_id, transcript_id, version, exact_excerpt, surrounding_question_answer, printed_page, line_start, line_end, pdf_page, coordinate_status, truncation_flag, ocr_flag, source_locator. No credibility or legal-relevance judgment. Use only these synthetic or firm-approved inputs and preserve source row/page IDs. Keep missing or conflicting evidence unresolved. Return a draft for the named reviewer. Do not send messages, connect accounts, change records or permissions, move money, or make legal decisions. Keep source files unchanged. List missing prerequisites before continuing. ``` Proposed / not tested · editorial prompt Published by Law Firm Agents. Proposed adaptation; not tested. Not legal advice.