How to Edit a Podcast in GarageBand on Mac

To edit a podcast in GarageBand for Mac, import each available recording onto its own track, align the conversation, and save a project. Split and remove unwanted sections, balance voice and music, then export and listen to the complete file. Keep the untouched originals so you can recover from an edit that cuts a word or moves a guest out of sync.

This is an audio-editing workflow for GarageBand on Mac. GarageBand calls timeline clips regions and calls a project a song; the same controls can arrange a conversation. iPhone and iPad have different interfaces. Apple's GarageBand for Mac guide covers tracks, regions, mixing and export.

1. Organize and align the recordings

1. Collect the sources. Keep the host, guest and any intro music as separate files where available. Label the originals and make a working project. A separate track can still contain another speaker through microphone bleed.
2. Import the audio. Start a Mac project and bring in the recordings, keeping each person on a separate track. The GarageBand import guide covers supported sources and checks.
3. Find a shared cue. Align a clap or an unmistakable word heard on both tracks. Listen for doubling while comparing the microphones. Check another shared cue near the end: independent recorders can drift, and a matched start alone does not prove the whole episode is in sync.
4. Audition tracks individually. Use solo or mute to find noise, overlapping speech and clipping, then restore the intended track states. Save the project and preserve the originals separately.

2. Cut mistakes without losing the conversation

First make a structural pass: remove a false start, a duplicated answer or a long interruption. Leave enough context for the exchange to make sense. For a middle cut in GarageBand's Tracks area:

1. Select the relevant region and move the playhead to the start of the unwanted passage.
2. Choose Edit → Split Regions at Playhead, then split again at the other boundary. Confirm which regions are selected before each split.
3. Remove the unwanted section. Decide deliberately whether to leave a gap, replace it with appropriate room tone, or close it. When moving material in a multitrack conversation, preserve the relative timing of every track that must remain synchronized.
4. Play from before the cut through the following sentence. Check that breaths, consonants and natural pauses survive, and listen for clicks or a sudden change in room sound.

Zoom in when placing a cut and inspect where the boundaries land. A split affects the selected material; it does not automatically make a clean join or keep other tracks synchronized. The GarageBand cutting guide explains trimming and splitting in more detail.

3. Choose a cleanup method for the actual problem

Match the problem to a first check

What you hear	First action	Limit
Steady hiss between words	Compare a quiet section with the original and test a suitable noise-reduction workflow.	Reducing hiss can soften consonants; a gate may leave noise audible under speech.
Music or another sound beneath speech	Look for a separate microphone track. If unavailable, compare a short voice-isolation test.	An estimated voice can retain music or lose words; it is not the original mic track.
Crackling on loud words	Compare the unprocessed source and another take.	Lowering the volume cannot restore peaks already clipped during recording.
Different host and guest levels	Adjust the separate track levels and listen through turn changes.	A loudness target alone cannot fix an inconsistent performance.

For one difficult passage, the voice isolator accepts a supported MP3, WAV, MP4 or MOV up to 50 MiB. Describe the audible speaker and compare Isolated sound with the source. AI WAV downloads require an account and sufficient credits. If you replace a passage in GarageBand, place it at the matching source time and mute the corresponding original to avoid doubled speech. This is a separate upload and download, not a GarageBand plug-in.

4. Balance the episode and export it

Play the host, guest and music together. Adjust individual track levels so the conversation remains understandable, especially when the intro enters or a guest speaks quietly. Compare at a comfortable listening level; avoid treating a taller waveform as proof of clearer audio. If music masks speech, lower or automate the separate music track and check the transition. The ducking guide explains that decision.

Choose Share → Export Song to Disk and select the format and export range appropriate for the destination. Apple's export example explains choosing an export range. Check any active cycle area or region selection so the entire episode, rather than a short section, is included. A saved GarageBand project is not the same as a finished audio file.

Open the exported file outside GarageBand. Listen to the beginning, edits, transitions and ending; verify duration and that no intended track was muted. For delivery measurements, see podcast loudness and true peak. The site's separate loudness meter supports short WAV files within its own limits; it is not a replacement for a full-episode measurement.

Common questions

Can GarageBand separate two people from one mixed recording?

Separate original microphone tracks offer independent level and editing control. Importing one mixed file does not create an individual track for each speaker. Source isolation is a separate estimate and may not distinguish overlapping, similar voices.

Can I use this workflow on iPhone?

The editing principles apply, but the controls and import limits differ. Apple's iPhone region guide documents its touch controls. When importing long audio on iPhone, check the song-section length and Apple's import instructions before assuming the whole file was included.

Does exporting publish the podcast?

No. Export creates an audio file. Review it, then follow your chosen host's upload and publishing workflow. Keep the editable project and source files for later revisions.