Edit a podcast on Mac.
Keep the conversation.

Start with the conversation.

Import the recording into a Mac project and listen to a short representative section. Check speech clarity, background sound, and whether speakers talk over one another. Keep a copy of your original recording.

Decide what this pass should do: remove a long gap, reduce a distracting pause, or prepare readable captions. Make one kind of change at a time. A separate pass for picture and titles keeps pacing decisions easier to judge.

Tighten pauses without losing meaning.

Choose the material to edit and ask for a narrow cleanup. The following is an example brief to adapt, not a guarantee that every pause will be interpreted correctly:

“Tighten long silent pauses in this selection. Keep deliberate pauses and do not remove spoken words.”

Review each proposed cut, especially around laughter, interruptions, and the beginning or end of an answer. A quiet moment can carry meaning. Undo a cut that makes the exchange feel rushed, and refine the instruction for the next pass.

For cleanup that depends on spoken content, verify the speech and selected source. Low sound level alone does not identify filler words or decide whether a breath should stay. Noisy recordings and overlapping speakers deserve a closer listen.

For a focused pass, follow the guide to removing long silences while reviewing each cut.

Make captions easy to follow.

Run Smart Captions on a clear audio selection and review the Creative Credit quote. Captions are placed as timed text clips on the timeline, where you can correct words, split or adjust sentences, and refine their timing.

Check names, specialist terms, and where one sentence ends and another begins. Use a readable font, sufficient contrast, and a text outline when the background changes. Inspect the final caption at the exact frame where it appears.

Caption generation is separate from automatic multicamera switching or identifying every speaker. This workflow does not promise either capability.

Follow the AI captions workflow for a separate generation, correction, and export check.

Listen to the edit, then watch it.

1. Listen across every cut for clipped consonants, unnatural gaps, and changes in background sound.
2. Check that answers still have context. A shorter episode should not change what someone meant.
3. Watch captions, titles, masks, and overlays together in the timeline preview.
4. Export a short test and play the actual output before exporting the entire episode.

If a result is wrong, keep the app version, timestamp, request ID if present, and exact instruction. Contact support with a small example you are allowed to share.

Choose the information your edit needs.

Available in Vibecut 0.4.3: open Analyze and choose the Podcast profile. It starts with Audio only for decisions based on speech and sound. Choose Video + audio when your brief depends on facial visibility, on-screen text, visual actions, or camera framing. Audio-only analysis cannot inspect those details.

Audio only costs 2 Creative Credits per started minute; Video + audio costs 5 Creative Credits per started minute. Fresh analysis uses the full duration of each unique source, rounded up separately. For example, a 90-second source costs 4 Creative Credits for Audio only or 10 Creative Credits for Video + audio. Trimming that source on the timeline does not reduce its analysis duration.

Review the source count, duration, and maximum quote before confirming. Changing the goal or mode may require fresh analysis. A successful analysis is charged even if you later choose a different editing direction; failed jobs release their reserved credits.

Understand analysis costs and plan allowances, or check the release plan. Basic timeline property changes do not require a full footage analysis first.