Assemble a clear episode, repair audio carefully, review the actual transcript and verify destination-specific exports before publishing.
Collect every speaker track, backup, running order and any approved music or ad material. Confirm episode and source versions, align tracks and keep untouched copies. Note missing audio, drift or consent/editorial restrictions before choosing the episode structure.
Select the intended content and remove errors without changing a speaker’s meaning. Preserve natural turn-taking and intentional pauses. Time the assembled content, intro, advertising and outro once in the runtime tool; use the actual pilot edit to estimate effort rather than a fixed hours-per-hour rule.
Diagnose clicks, steady noise, reverb and level differences separately. Test cleanup on representative quiet and loud passages, comparing at matched level. Trial a static compression curve if useful, then listen to real transient timing, breaths and pumping before accepting the settings.
Obtain or prepare the transcript through the chosen transcription workflow and check names, numbers, speakers and timestamps against the recording. Count actual reviewed text in Word Counter. Use the transcript-budget tool only for estimated speech workload and separately billed audio duration; it cannot transcribe the episode.
Measure full-program loudness and true peak in audio software against the selected destination guidance. Compare constant gain with an explicit peak ceiling, then remeasure after final processing and encoding. Choose codec and bitrate for the actual channel layout and host requirements; retain an editing master.
Play the final encode from start to end, checking transitions, music, ads, silence, sync and the ending. Match the title, notes, links, transcript and chapter timestamps to that exact version. Save the approved file, source project and correction record before handing it to publishing.
Write and download the task brief, measured results and handoff decisions as Markdown.
Add actual planned content, introduction, advertising and outro durations once.
Inspect the static hard-knee curve with explicit makeup gain before auditioning the real processor.
Compare the measured sample or true peak with an explicit ceiling; keep crest factor separate.
Count words in the actual reviewed transcript or script text that you paste into the tool.
Estimate transcript workload from speech minutes, and price billed audio minutes separately.
Compare a constant loudness adjustment with the measured true peak and an explicit delivery ceiling.
Estimate delivery or backup size from the actual encoded bitrate or PCM format.
Loading your checklist…
Apple distinguishes RSS audio from subscriber-audio delivery and gives channel-dependent bitrate guidance. Check the intended delivery path directly.
Noise profiling and reduction require a representative sample and listening for removed wanted content.
Keep the following evidence with the actual version used for this task.
| Record | Include | Check |
|---|---|---|
| Editorial | Approved assembly, duration and restrictions | Compare meaning with original takes |
| Text | Reviewed transcript, names and timestamps | Match the final audio revision |
| Export | Codec, channels, peaks and loudness | Listen after encoding |
Track editing, proof-listening and correction time separately for your own format.
A late cut can invalidate every later chapter timestamp.