Cut

TunaCut Editor

A real editor with multicam switching at its centre. The top lane of the timeline is not a video track — it is a contiguous record of which camera is on screen, and editing is mostly the act of changing your mind about that.

The TunaCut editor: program monitor, three camera tiles, the camera panel and a timeline with the camera-view bar, speaker ribbon and two waveform lanes

Program monitor, every angle live beside it, an inspector rail on the left and the timeline underneath. There is no Save button — the edit is written after every action.

The timeline

Six lanes, and only two of them are clips

Camera View — the switch lane. Every section points at a camera; a split makes two, and a skip removes time from the finished video without deleting anything.
The speaker ribbon — one colour per microphone, showing who was talking, measured from the levels rather than guessed from which mic a line came off.
A waveform lane per microphone, each with mute, solo and gain, and each renameable to the person sitting behind it.
Music, Text and Media rows for everything laid over the top — a track with a drawn volume line, titles, and images or video inserts that can override the timeline underneath.
The ruler carries chapters and the red and green marks you pressed while recording, with their notes.

The cameras agree with each other

Sync is measured from the audio and shown as a number you can argue with — nudge it by a frame or ten, or type the offset in seconds.

Match the cameras pulls three angles into one look, which matters most when they are different bodies with different sensors. Framing nudges, a fill-the-frame control and named camera profiles that carry from one shoot to the next sit beside it.

Who is talking, and what they said

Measure is local and free: the speech envelope, the speaker track, and every pause classified — a breath, a gap between sentences, or dead air worth removing.

Transcribe adds the words, with each word boundary snapped to the nearest real speech edge from the waveform rather than trusted from the transcript, because transcription drifts by a few frames and a cut lands on a frame.

The Voices panel showing microphone count, pause analysis and a transcript with timecodes

The Voices panel: analysis summary, who-is-talking legend, pause counts and the full transcript

A measured session: how many microphones, how much overlap, how many sentences, and every pause counted and sorted by kind.

Cuts

One take. Every cut.

A Cut is a whole finished version of the recording, kept as its own project file beside your rough pass. Generating one is safe because the Original is never modified — the worst outcome of pressing the button is a file you delete.

The Cuts panel listing the Original and a generated Distributed Cameras cut

Distribute Cameras writes a full switching pass into a new Cut, and tells you why: how many changes it made, how many were speaker changes and how many were a shot going stale.
Shorts pulls the moments worth lifting out into their own vertical Cuts, with the on-screen text already placed.
Each Cut can be renamed, duplicated, reframed, marked long or short, or deleted — and opening one is a timeline swap, not a reload.

Vertical without re-shooting

Turning 16:9 into 9:16 is a framing decision, so TunaCut makes it per shot. Where one person is talking the crop follows the head; where two are in frame the whole picture is kept and the space above and below is filled with a blurred copy of it.

Sound that is finished, not just present

Normalise the microphones to a target loudness — −16 LUFS for a podcast, −14 for YouTube — even out loud and quiet delivery, reduce the bleed between two microphones in one room, and stop anything clipping. Every one of them is reversible in a click.

Built for hands on a keyboard

Play, shuttle forward and back, step a frame, jump to the next camera change or the next mark, split, skip, set in and out, insert media, add a title — all rebindable, keyboard and mouse alike.

A Blackmagic Speed Editor is supported directly: jog, shuttle and scroll wheel modes, the camera keys lit to show which angle is live, and an exportable key map.

Export

Four sizes, three intentions

720p to 4K, and a quality choice phrased as what the file is for: keeping and re-editing, uploading, or sending to someone. Software encode for the best picture, hardware when you would rather have it now.

You can keep working while it runs. And the line at the bottom of the dialog is the whole philosophy of the application: your recordings are never modified.

The export dialog: destination, size, quality, encoder and progress

The first time you open a session, it walks you through it

Five steps: build the preview media, mark what is left out, say who is on each microphone and which camera is theirs, match the cameras to each other, and set the reference head for vertical clips. Then it offers to cut the cameras together.

Read the walkthrough What happens next →