Review speaker turns before cutting an interview
Make a source-timed paper edit for a short two-person interview without turning uncertain speech into a confident quotation.
A source-timed speaker log, a proposed 30-second paper edit and an unresolved-words list for a human editor.
Original FreeStack editorial workflow, checked 12 November 2026. The illustrated rows are invented planning examples, not transcription output, tested timings or customer results.
About this guide & its examples
Use a short recording you have permission to process, or make a fictional practice recording.
Illustrations are FreeStack concepts, not provider screenshots. Check the free limits below and test your own files before delivery.
Free boundaries before you start
What’s free, whether you need an account, and when we last checked.
Audacity · checked 2026-11-12
Ongoing free GPL desktop audio editor for Windows, macOS and Linux. Direct download is available; MuseHub is an optional distribution route. Paid cloud storage and third-party plugins are separate from free local recording/editing/export. Some media formats need additional components; imported audio rights remain separate.
Core local editing does not need a hosted account; cloud saving/sharing has separate account terms.
LibreOffice · checked 2026-11-12
Free desktop office suite with no licence fees for personal or business use; MPL 2.0 core with separately licensed components. Writer supports local editable documents and PDF export; no subscription or numeric export quota indicated. Embedding permitted fonts can improve source-file portability and increases file size. Professional support and third-party templates are separate.
No hosted account is part of the described local document workflow. The checked FAQ offers direct free downloads and optional donations.
A small example to aim for
A speaker label is a place to check, not a name.
Source time → voice label → context check → approval.
Invented planning rows only. Not a Scribe transcript, tested recording or customer result.
Illustrative example · adapt it to your project.
Jump to a step
Define the edit before opening a service
Choose one question and a roughly 30-second answer-led edit. Keep the original recording unchanged. Note its filename, duration and revision in a LibreOffice Calc sheet. Make columns for source start, source end, speaker, words, confidence, proposed use and reviewer. A paper edit is a plan, not a finished video.
Listen without changing the source clock
Open a working copy of the audio in Audacity. Listen to the whole exchange once, then select short ranges to replay difficult words. Do not delete pauses or change speed in the source copy used for timing. If your recording has separate channels, inspect them without assuming each channel always belongs to one person.
Separate the voice label from the person
Enter Speaker A and Speaker B at the points where the audible voice changes. Add a name only after checking the recording context or confirming it with the project owner. Mark overlapping speech and uncertain words explicitly. Do not fill a gap using what would make the sentence more convenient for the edit.
Build a proposed sequence from intact thoughts
In Calc copy the candidate ranges into an edit-order section, retaining their original timestamps and context notes. Choose an opening statement, supporting detail and ending. Sum the proposed range lengths, including any planned pauses. Ask whether removing the interviewer’s question changes the meaning. Reject a cut that makes a hesitant or conditional statement sound certain.
Listen back to the chosen ranges
Replay every chosen range in Audacity with a little surrounding context. Compare wording, voice label and boundaries against the sheet. Record who checked it and what remains unresolved. A second reviewer should check sensitive quotations. Keep proposed and approved rows separate; do not claim approval because the spreadsheet is complete.
Hand off the plan and save the kit
Deliver the source reference, paper edit and unresolved list to the editor using approved storage. Keep a working ODS file and export a PDF for easy review, then inspect that PDF. Use Save this toolkit for the two tool links. FreeStack editorial upgrade triggers: recurring transcription volume, a verified language requirement, approved team review or a justified data-retention need. Evaluate that requirement before buying a membership.
Your workflow, connected.
Follow the steps and carry the output into the next stage. These are suggested handoffs, not automatic integrations.
Provider information
Check current plans and documentation before committing to a workflow.
A couple of useful questions
What stays free?
Audacity’s local desktop editor is ongoing free GPL software; LibreOffice is MPL 2.0 software with component terms. Local replay and spreadsheet work have no provider minute, export-credit, watermark, storage or seat quota. Your hardware, disk and media rights remain separate; paid cloud services and plugins are optional.
Do I need an account?
Core local listening and spreadsheet work do not require provider accounts. Downloads, optional cloud storage and delivery services can have separate requirements. Signed-in FreeStack saves are private account stacks; guest saves stay in that browser.
Is this a subtitle workflow?
No. This produces a source log and proposed edit order. Subtitle timing, readable line breaks, rendering and destination compatibility are separate jobs. The existing caption guide covers a portable SRT workflow.
Can I start with AI transcription?
Yes, if the recording may be sent to the selected service and its current terms fit the project. Treat every word and speaker label as a draft. Keep the source clock, replay chosen quotations and retain uncertainty. Do not assume free credits include the required model, retention settings or commercial licence.
What would you try?
Share your result or ask a specific question about this workflow.
Join the conversation