Speaker Cut. Download free
The Speaker Cut 2.0 guide

From the shoot to a finished Final Cut edit.

Bring in your cameras and mics, tell Speaker Cut who's who, and it cuts the whole episode to whoever is talking. Watch it, switch it live if you disagree, and send it to Final Cut Pro in one import.

~10 min end to endSpeaker Cut 2.0Final Cut Pro 12+Apple silicon · macOS 15.6+
One recording?No separate mics, just a Zoom, StreamYard or single-camera recording? Use the Highlight Cut guide instead: Speaker Cut tells the voices apart for you.
Part 1

Start an episode.

Home has four ways in. Pick the one that matches how you recorded, or just drop your footage anywhere on the Home window and Speaker Cut opens the right one.

Before you start: keep the drive with your footage connected the whole time. Speaker Cut never copies or uploads your media; it reads it where it lives.
1A · From clips

Build a multicam from clips

For cameras plus separate recorders: lavs, wireless packs that record onboard, a Zoom or field recorder, a StreamYard program track.

  1. Drop the whole shoot

    On Home choose Build a multicam from clips, then drop the shoot folder, or use Choose a folder… / Choose files…. Each camera or recorder becomes one angle. Recorders that split a long take into ~30-minute files are stacked into one angle automatically.

  2. Check the angles

    Angles are named from your files and folders; click a name to change it. Click an angle's file line to reorder, add or remove pieces. Add a missing angle at the bottom, or drop more files anywhere on the window.

  3. Mark the master with ◆

    The master is the one feed that hears everyone: a program mix, a room mic, or a wide camera that hears the whole table. Speaker Cut guesses it from the names (program, mix, room, wide); click the ◆ on another angle to change it.

  4. Sync by sound

    Click Sync by sound. Speaker Cut listens to every angle against the master. No timecode needed. Each angle gets a status:

    MasterEverything syncs to this.
    Locked · −1.2 sIn sync, with its offset.
    Low confidenceWorth a check.
    No lockCouldn't hear a match.

    All locked? You'll see “Synced. All angles line up…”. If an angle says No lock, make sure it recorded the same conversation (camera mic on) or pick a master that hears that person, then Sync again.

  5. Next: who's who

    Click Next: who's who → and carry on with Part 2.

Speaker Cut — New multicam
The New multicam screen: drop the whole shoot here, then lanes with a ◆ master column and a Sync by sound button.

Build a multicam: drop the shoot, mark the ◆ master, Sync by sound.

1B · ATEM ISO

Import an ATEM recording

For shoots recorded on an ATEM Mini Pro ISO or Extreme ISO, with or without wireless mics that record onboard.

  1. Choose the .drp

    On Home choose Import an ATEM recording and click Choose the .drp. It sits next to the Video ISO Files folder. You can also drop it on the window.

  2. Pick the take

    Each take shows its length, start timecode and number of live switches. The longest take with cameras is picked for you.

  3. Check the cameras

    Every camera is already in sync because the ATEM recorded them together. Inputs that recorded black are left out; click a camera to leave it out yourself. The ATEM's own live cut (PGM) is kept for reference.

  4. Choose the master audio

    The sound that hears everyone: what the ATEM sent out live, one of its mic inputs (for example the input carrying your mix, like MIC 1), or a camera's sound. It also lands in Final Cut as a muted backup lane.

  5. Add your mic recorders

    Click Add a folder for each wireless pack or field recorder (or Add files). Add the whole folder: Speaker Cut finds where each file sits in the take, joins the 30-minute pieces, and sets aside files that belong to a different take.

  6. Line everything up

    Click Line everything up. The summary shows each recorder as Locked (or Partial if it covers less of the take) and lists any files set aside, with the reason. The aligned audio is saved in a Speaker Cut Audio folder next to your recording.

  7. Next: who's who

    Click Next: who's who and continue with Part 2.

Tip: keep each wireless pack in its own folder named after the person, like H1 - Matthew. Lanes take the folder's name, so who's who is a two-click job.
Speaker Cut — Import an ATEM recording
The ATEM import screen: Every camera, already in sync. Choose the .drp.

Pick the .drp, the take, the master audio and your mic folders, then Line everything up.

1C · Already synced

I already have a multicam

Synced it in Final Cut already? Bring it over and go straight to who's who.

  1. Export it from Final Cut

    Select the multicam clip in the browser, choose File › Export XML… and save it. (Export the multicam clip itself, not a project.)

  2. Drop it on Speaker Cut

    Choose I already have a multicam and drop the .fcpxmld, a .zip of it, or the Info.fcpxml inside.

  3. Next: who's who

    Continue with Part 2. Your original multicam is kept untouched and can go into the export package as Source.fcpxml.

Speaker Cut — Bring your Final Cut multicam
The I already have a multicam screen: drop it here, with the three Final Cut export steps on the side.
Part 2

Tell it who's who.

One screen, shared by all three ways in. Speaker Cut needs to know whose mic is whose (that's how it hears who's talking) and what each camera shows.

  1. Pick the show

    Choose the episode's Show. The cut feel and auto-mix start from that show's settings, and if the show knows its hosts' voices, Speaker Cut listens for them and offers “Recognized by voice…” with Use these names.

  2. Name the people

    Under People, name everyone who talks. Add a person adds another. Each person needs a mic; a camera is optional.

  3. Mic: whose audio is each lane?

    In the Mic column, set each recorder to its person (Matthew's mic), the feed that hears everyone to Master audio, and anything else to No mic. Not sure? Click Listen to hear a few seconds of that person.

  4. Shows: what does each camera show?

    In the Shows column, set each camera to a person, a pair (a two-shot, like Felipe + Ben), All speakers (wide) or Don't show. One camera must be the wide.

    • Two cameras on one person? One becomes their A-cam; the others punch in now and then as B-cams for variety.
  5. Continue to cut

    Continue to cut lights up once everyone has a mic and one camera is the wide. The status line tells you what's missing.

A four-person panel: three wireless mics, a camera per host, a two-shot and the wide.

Part 3

It's already cut. Watch it.

The Cut tab opens with the whole episode cut to whoever is talking. Press play: the switching you see is exactly what lands in Final Cut.

  1. Play it

    Space plays and pauses; ← → step a second (⇧ for five). Every camera is on screen and the mics play mixed the way Final Cut will hear them.

  2. Disagree? Switch it live

    While it plays, press a camera's number, 1–8, like a switcher. Your switch replaces the cut from the playhead to the next change. ⌥+number switches the whole shot; ⌘Z undoes.

  3. Tune the feel

    In the inspector, Cut feel has Pacing (Calm, Balanced, Lively), the shortest shot, whether to Stay on speaker through a quick “yeah”, how often to Cut back to the wide, and Go wide on cross-talk. Every change re-cuts the whole episode in a moment, and your own switches are kept.

  4. Auto-mix the mics

    Auto-mix opens each mic when that person talks and turns it down when they don't. Choose how hard (Gentle to Gate), set the fades, or drag a ◆ on a mic lane to reshape it by hand.

  5. Save it for your show

    Happy with the feel? Save for ‹your show› makes it the default for every new episode of that show.

Overview shows each person's share of the screen next to their share of the talking, plus the shot count and average shot, so you can see at a glance if someone is under-covered.
Speaker Cut — Cut

Real playback: four automatic cuts in fifteen seconds. Press 1–4 to take over.

Part 4 · One & Pro

Transcript, notes & highlights.

With Speaker Cut One or Pro, the same episode gets a word-by-word transcript, show notes, chapters, highlights and cold-open hooks. The transcript runs on your Mac; only its text goes to the AI.

Transcript

  1. Transcribe

    On the Transcript tab click Transcribe. Each word is credited to whoever said it, even when every mic hears the whole room.

  2. Fix a speaker

    Click a speaker's name ▾ and pick Who said this?. Captions, the export and the AI all use the fix.

  3. Find and download

    Search with ⌘F, click any word to jump the video there, and Download Captions .srt, Transcript .txt or a Word doc.

Speaker Cut — Transcript
The Transcript tab: speakers on the left, the word-by-word transcript, and the video following along.

Highlights & hooks

  1. Find highlights

    On Highlights, optionally say what to look for (“stories about the early days”), then Find highlights. Each highlight is a complete moment, stitched from anywhere in the episode, ending on its payoff.

  2. Read why, then refine

    Open a highlight to see its pieces and why it was picked. ↻ Re-roll and tell the AI what to change (“tighter”, “add the part about pricing”), or leave it empty for a different pick.

  3. Ask for more

    Find another highlight lets you pick a flavor: A story, Advice, A hot take, Something funny… and preview each one before adding it.

  4. Build the reels

    + Add to reel and + Add to cold open build the Highlights reel and the Cold open; reorder them on the Reels tab.

Speaker Cut — Highlights
The Highlights tab: chapters on the left, a selected highlight with its stitched pieces, and highlight cards below.
Part 5

Send it to Final Cut.

One import brings everything over, built from your original footage (never the light playback copies).

  1. Check what lands

    The Export tab lists it all: the full cut, the Highlights reel, the Cold open, and a clip per highlight and hook. Untick anything you don't want.

  2. Choose the audio

    Auto-mix keyframes, Level matching and a muted Backup mix lane; audio roles One per person or One Dialogue role.

  3. Send to Final Cut

    Click Send to Final Cut. Final Cut opens and asks which library to add the event to. Each mic arrives on its own lane, keyframes and all. (If Final Cut doesn't open, import the saved file with File › Import › XML….)

  4. Or build the package

    Build package (.zip) saves the Final Cut file with whatever you tick:

    FileWhat it is
    Final Cut.fcpxmlThe import, always included
    Notes & Highlights.pdfSummary, show notes, highlights, hooks
    Transcript.pdf / .srtEvery word by speaker; captions
    YouTube chapters.txtReady to paste into the description
    Camera stillsA still from each camera
    Source.fcpxmlYour original multicam, untouched
    READ MEHow to import, for whoever opens it

    It saves next to your footage when that drive can be written to; Change… picks another folder for this episode.

Needs Final Cut Pro 12 or later. Speaker Cut writes FCPXML 1.14; Final Cut 11 and earlier can't open it. Update Final Cut from the Mac App Store.
Speaker Cut — Export
The Export tab: audio options, what lands in Final Cut, the package file list, and Send to Final Cut.
Good to know

Common questions.

Do I need timecode?

No. Speaker Cut syncs by sound, listening to each angle against the master across the whole session. If your cameras carry timecode, it's kept.

My audio is on separate recorders, not the cameras. Does that work?

Yes, that's what it's built for. Add the recorders as their own angles (or their folders on an ATEM import), then in who's who set each one as that person's mic and their camera in the Shows column. Speaker Cut hears who's talking from the clean mic and cuts to their camera.

An angle says “No lock”. What now?

Speaker Cut couldn't hear enough of the same conversation between that angle and the master. Check the angle recorded sound, and pick a master that hears everyone (a program mix, room mic or the wide). Then Sync again; it takes seconds.

Can I change the cut after it's made?

Any time. Switch shots with 1–8 while it plays, change the pacing or the wide, and the episode re-cuts while keeping your switches. Everything you change goes to Final Cut.

Do I have to build the multicam in Final Cut first?

No. Drop the raw clips or the ATEM .drp and Speaker Cut builds it. If you already synced one in Final Cut, export it and use “I already have a multicam”; both get the same cut.

Is my footage uploaded anywhere?

No. Sync, the cut, transcription and auto-mix run on your Mac. With One or Pro, only the transcript text is sent to write notes and highlights.

What do I need to run it?

An Apple silicon Mac on macOS 15.6 or later and Final Cut Pro 12 or later. See the full system requirements.

What are we cutting today?

Download Speaker Cut, drop in your footage, and watch the first cut play. The editing engine is free forever.