Record a podcast with separate tracks for every voice

A podcast recorded as one mixed file is fine until something goes wrong in it. A guest who sat too far from the microphone, a cough over someone's best line, one cheap microphone that hums: in a single mixed file, fixing any of these touches everyone at once. When you record a podcast with separate tracks, each voice is its own file, so you fix the one track that needs it and leave the rest alone.

This guide explains what separate tracks give you, how to record them in Chrome with Cheap Recorder for guests in the room and guests on a call, one worked example from start to edit, and the limits worth knowing before the episode that matters.

What separate tracks let you do

Balance voices. Make a quiet guest louder without making the loud host louder too.
Cut without collateral. Remove a cough, a door or a "sorry, go on" from one person while the other keeps talking.
Rescue a bad microphone. Clean up the one track with the hum, rather than the whole episode.
Keep a safety net. You still get the normal mixed recording, so a quick episode needs no editing at all.

What you need

Google Chrome on a Mac, Windows or Linux computer.
One microphone per person in the room, all plugged into the same computer. Wired or USB microphones work best. Avoid Bluetooth headsets as microphones, because they drop to a low-quality mode when used that way.
Headphones for everyone, so no microphone picks up sound from a speaker.
For remote guests, a Google Meet or Microsoft Teams call open in a Chrome tab.
Free disk space. Separate files take more room than a single MP4, and every file is written to a folder on your own computer.
Your guests' agreement to be recorded, and a check of the rules where you and they are.

How to record a podcast with separate tracks in Chrome

Open Cheap Recorder in Chrome and sign in. Under Where to save, press Choose folder…. Each session gets its own folder inside it, named with the date, the time and the session name.
Open Options and tick Also save every source separately. If you use studio microphones in a quiet room, also tick Record the microphone untouched, which turns off echo, noise and volume correction. Leave it unticked on laptop microphones or in a noisy room.
In the Microphone tile, open Change or add microphones and tick every microphone you want recorded. Each shows "wired" or "bluetooth" beside its name. If one is missing, plug it in and press Rescan. Some plans limit how many microphones you can record at once, and the start bar tells you if you tick too many.
For a remote guest, press Pick the call tab… in the Call audio tile, choose the call's tab in Chrome's window and tick Share tab audio.
For a sound-only podcast, switch off Record my webcam and do not pick a screen; the mixed recording is then an M4A audio file. For a video podcast, leave the webcam on or choose Webcam only.
Name the episode in Name this recording and press Start recording. A level bar next to each microphone moves with its sound: check every bar moves when its person speaks.
Press M or Add marker at each new topic, or when someone fluffs a line. The marker times are saved in markers.md.
Press Stop and finish. Every file is already in the session folder.

Before a real episode, record thirty seconds with everyone saying their name, stop, and open the session folder to check there is one microphone file per person. The full guide is Record a podcast with every voice on its own track.

What lands in the folder

You keep the usual mixed recording with every sound together. Next to it, a folder called Separate sources holds each source as its own WebM file:

one audio file for each microphone you ticked, named after that microphone and starting with audio_;
audio_meeting_far_end.webm for the other side of a call, if you picked one;
webcam.webm if your camera was on, and screen_with_mixed_audio.webm when a layout puts the screen and camera together;
helper files: manifest.json (a list of what was recorded), markers.md (your marker times) and finalise.sh.

Microphones with clearly different names are easiest to tell apart in the folder, because the files take their names.

A worked example: two hosts and a remote guest

Say you and a co-host sit at one desk, each with a USB microphone plugged into the same laptop, and your guest joins from a Google Meet call open in a Chrome tab. The show is sound only.

Set up. Choose a folder, tick Also save every source separately, tick both USB microphones, pick the Meet tab with Share tab audio, switch off the webcam and name the recording "Episode 12".
Test. Record thirty seconds where each of you, and the guest, says their name. Open the folder: you should see two files starting with audio_, one per USB microphone, plus audio_meeting_far_end.webm for the guest.
Record. Start again, clap once just after you press start, and press M at each new topic.
Edit. Load the three WebM files into your audio editor and line up the clap. The guest is quieter than you two, so raise only their track. Your co-host coughed over the guest's best answer, so cut the cough from the co-host track alone. Use the times in markers.md to jump to each topic.

If the episode needs no edit after all, the mixed recording beside the folder is ready to publish as it is. For more on capturing both sides of a Meet call, see how to record a Google Meet call with everyone's voice.

Limits worth knowing

Remote guests share one track. The call's sound comes from one Chrome tab, so everyone on the far end of the call is in the same file. If you want two remote guests on separate tracks, each has to record themselves.
Microphones in one room hear each other. Each file holds what that microphone picked up, so a loud neighbour is faintly audible on your track. Sit a little apart and speak close to your own microphone.
A microphone can drop out. If one disconnects, a message names it and its file stops at that point. The others carry on.

Editing the tracks

The recorder opens the sources one after another when you press start, so the files can begin a fraction of a second apart. The old studio trick fixes it: clap once, sharply, just after you start. In your audio editor, line up the clap on every track and they stay together from there.

If your editor does not open WebM files, the session folder has a finalise.sh script that turns each microphone's file into a WAV file. It is a bash script and needs the free ffmpeg program installed on your computer.

If Live transcript is on, the page shows the transcript after you stop. On plans that include speaker detection, each line is labelled with the source that was loudest while it was spoken, and you can rename those labels in the Speaker boxes to your guests' real names. On plans that include cutting words from the transcript, the same transcript helps you cut filler words.

Frequently asked questions

Can two remote guests each get their own track?

Not from one call tab. The call's sound is one source, so everyone on the far end shares a file; if you need them apart, each guest records themselves.

Why are my tracks slightly out of step?

The sources open one after another when you start, so the files can begin a fraction of a second apart. Clap once after you start and line up the clap in your editor.

My editor will not open WebM files. What now?

Run the finalise.sh script in the session folder. It turns each microphone's file into WAV, and it needs ffmpeg installed.

Do my recordings leave my computer?

The files are written to the folder you picked and are not uploaded, unless you use Convert. While the live transcript is on, Chrome sends your microphone audio to Google to turn it into text.

Get started

The free plan needs no card. Try Cheap Recorder in Chrome, tick Also save every source separately, and your next episode arrives with every voice on its own track, on your own disk.