Create your first rough cut
Every control on the New Rough Cut screen, and the safest defaults for a first successful result.
Choose New Rough Cut in the sidebar, drop in a video or an audio file, and leave the recommended defaults: Standard recording environment, On-device transcription, Bad-take removal on, Second Look and the annotated review timeline off, Auto video type, Standard edit intensity, Auto-detect language, and no video goal. Click Process and leave your source file where it is. RoughCut never copies or re-encodes your footage — the exported timeline references the original in place, and an audio-only recording produces an audio timeline with the same cutting controls. The screen remembers the settings you used last time, never the media or the goal, and Reset restores the defaults. A completed rough cut spends one of your five included runs, which include every Pro feature, but only when it actually finishes: failed and cancelled jobs stay free, and Transcribe Only never counts. Every other control on the screen has its own page for when the defaults are not right for your recording.
On this page
Watch this step in the getting-started video, then follow the written steps below:
Prerequisites
- The speech model installed, unless you are using Pro Cloud transcription. See Install the local speech model.
- An AI provider key only if you want bad-take removal. Silence-only cuts need no key.
- A supported video or audio file. See Choose video and audio inputs.
Steps
- Choose
New Rough Cutin the sidebar, underStudio. On anything but your first job the screen opens with the settings you used last time; clickResetif you want the defaults below. - Drop a video or an audio file onto the drop zone, or click
Browse Video…orBrowse Audio…. The zone readsDrop a video or audio file here, andDrop to replace the video or audioonce a file is chosen. A WAV, MP3 or M4A on its own is enough: RoughCut builds an audio timeline from it. - If you have both a video and a separate, cleaner microphone track, add both. As the screen says,
With both, the audio is used for analysis and the timeline references the video. - Leave
Recording environmentonStandard. That is −25 dB and suits typical home or office recordings. - Leave
TranscriptiononOn-device. Nothing is uploaded for transcription on this path. - Leave
Bad-take removalon if you have a provider key. Turn it off for a silence-only cut that needs no key. The caption under it is the cost rule in one line:AI usage is billed separately. Cost depends on transcript length, model and extra passes.— theAPI cost guidebutton beside it opensHelp → API Costs. - Leave
Second LookandAlso export an annotated review timelineoff for a first run. Both are Pro options: the first adds another AI call, the second adds a second FCPXML for reviewing the cuts inside Final Cut Pro. - Leave
Fine-tunecollapsed. Its defaults are 0.5 s minimum silence,Standardaggressiveness,Autovideo type,Standardedit intensity,Auto-detectlanguage, and no video goal. - Click
Process. The job's progress screen opens straight away, and the button stops accepting clicks so a second press cannot create a duplicate job. - Keep the source file where it is until you have exported and imported the result.
Expected result
The job moves through its stages and finishes with a review list of proposed cuts, then a Your rough cut is ready banner with an Export for Final Cut Pro… button — or Export Both Timelines… when the job also produced the annotated review timeline. During the AI stage the progress view shows an animated indicator and elapsed time instead of a percentage; see Understand job progress.
The controls, in one place
| Control | Default | What it changes | Pro |
|---|---|---|---|
Recording environment | Standard (−25 dB) | How quiet counts as silence. Quiet room −35, Noisy −20, Custom dB, or Smart | Smart only |
Transcription | On-device | Whether audio is transcribed locally or by OpenAI | Cloud only |
Bad-take removal | On | Whether the AI editor pass runs at all | No |
Second Look | Off | A second AI read of the surviving footage | Yes |
Also export an annotated review timeline | Off | Adds a second FCPXML in which nothing is deleted — every decision a colour-coded clip for review in Final Cut Pro | Yes |
Include cut silences as clips | Off | Sub-toggle of the above: keep removed silences on the review timeline as RC Silence clips | Yes |
Min silence (sec) | 0.5 | Pauses shorter than this survive | No |
Bad-take aggressiveness | Standard | How boldly the editor removes suspected bad takes | No |
Video type | Auto | Explainer / talking head cuts weak sentences harder; Screen demo keeps narrated on-screen steps | No |
Edit intensity | Standard | How much the video should shrink: Light keeps most takes, Heavy condenses whole passages | No |
Language | Auto-detect | Transcription language | No |
Video goal (optional) | empty | Editorial context, max 500 characters | Summarising only |
Second Look is disabled unless Bad-take removal is on, because it re-reads the result of the first AI pass. The same applies to Also export an annotated review timeline: with no editor pass there are no decisions to annotate. Both carry a PRO badge, and turning either on without Pro access opens the Pro paywall and leaves it off. The review timeline has its own page: Review every cut in Final Cut Pro.
Video type and edit intensity
Two pickers inside Fine-tune tell the editor what kind of video it is cutting and how hard to cut. Both are core controls, not Pro.
Video type offers Auto, Explainer / talking head, and Screen demo. The help text under it:
Screen demos keep narrated on-screen steps; explainers cut weak sentences harder. Auto
guesses from the transcript.
Choose Screen demo when the words describe what is happening on screen — a step you narrate while you click is not a tangent, and this setting stops the editor treating it as one. Choose Explainer / talking head for a piece to camera where a weak sentence can go. Auto reads the transcript and picks.
Edit intensity offers Light, Standard, and Heavy:
How much this video should shrink. Light keeps most takes; Heavy condenses whole passages.
Standard matches the owner's median edit.
It sits next to Bad-take aggressiveness, and the two answer different questions: aggressiveness is how boldly the editor acts on a weak signal, intensity is how much material it should aim to take out. Change one at a time and judge the result on the same recording.
The screen remembers your settings
New Rough Cut opens with the settings you used on your last job, and says so in the caption under the controls:
Your settings are remembered for next time. Reset restores the defaults.
What is remembered: the recording environment, the transcription source, the toggles, and everything under Fine-tune except the goal. What is never remembered: the media you chose, and the Video goal text — a goal describes one video, so the field starts empty every time. A Pro option is not pre-selected while Pro is unavailable to you, even if you used it on an earlier included cut.
Reset does two things at once, as its tooltip says: it clears the media and returns every control to the defaults in the table above, and the remembered settings are forgotten. The defaults in this article are what a first-ever job starts with, and what you get back after Reset.
Your original file is never touched
RoughCut states this on the screen itself: files are never copied, and the FCPXML references your original file in place. Nothing is transcoded, nothing is re-rendered, and no second copy of your footage is created. What RoughCut produces is a set of cut decisions.
Audio-only recordings
A podcast episode, a voice memo or a narration track can be cut without any video. Drop the WAV, MP3 or M4A on its own, or choose it with Browse Audio…, and process as usual. The app's Help puts it this way: Process creates an audio timeline you can import into Final Cut Pro, with the same cutting controls and optional review timeline. The exported FCPXML references the audio file in place, exactly as it would a video, and the review list, Recompute Cuts and the Pro annotated review timeline all work the same way.
How the trial counts
While you have not purchased the unlock, the screen shows <n> free cuts left. Five complete rough cuts are included, and while cuts remain a second line reads Every Pro feature is included in your first 5 cuts (<n> left). — Smart Silence, Cloud transcription, Second Look, the annotated review timeline, and goal summarising all work in those five without a subscription. See Trial, core unlock, and RoughCut Pro.
A run is spent only when a job actually reaches a completed state. Specifically:
- Failed jobs do not count.
- Cancelled jobs do not count.
- Transcribe Only never counts, and is never gated by purchase at all.
The counter lives in the macOS Keychain, so it survives deleting and reinstalling the app, and nothing in the app lowers it.
Common mistakes
- Adding a separate audio file that is not synchronised with the video. When both are present the audio track is used for analysis and transcription; if it drifts from the video, every cut time drifts with it.
- Moving or renaming the source file before importing. Final Cut Pro then reports the media as offline.
- Expecting the screen to start from the defaults. It starts from your last settings. Click
Resetto go back to the defaults. - Turning on
Second Lookfor a first job. Learn what the standard pass does before adding a refine pass. - Setting a custom dB value before trying the presets. The three presets cover most recordings.
- Expecting
Processto work with no model and no Pro. On-device transcription needs the model; the button stays disabled and tells you so.
Troubleshooting
Process disabled? Check in order: a video or audio file is selected; no red validation message is showing; the speech model is ready (or Cloud is selected); you have free cuts remaining or have purchased the unlock.
If the job fails, read the message before deleting anything — it names the cause. See Troubleshooting by symptom.

Second Look and Also export an annotated review timeline turned on.