Why audio handling decides the quality of your edit
Ask any working editor what separates a polished video from a rough one and the answer rarely has anything to do with fancy transitions. It comes down to whether the audio behaves. Dialogue that jumps in level between takes, camera hiss sitting under a music bed, a music track that suddenly cuts because a clip was moved three frames to the left — these are the details viewers notice without knowing they notice them.
Two operations sit at the center of almost every edit session: getting rid of audio you do not want, and putting new clips where you do want them. They sound trivial. They are not. Most timeline chaos — misplaced clips, accidentally overwritten footage, out-of-sync dialogue, music that starts a beat too early — traces back to a handful of misunderstood buttons in the Edit page of DaVinci Resolve.
This guide walks through the whole chain: how the Edit page is organized, how to detach audio from video, how to delete or mute it cleanly, how to add fresh clips without destroying what is already on the timeline, how to sync new material to existing footage, and what to do afterward with room tone, ambience, and voice. It also covers the mistakes that catch people again and again, and how to recognize them before they cost you an afternoon.
The Edit page layout you actually need to know
DaVinci Resolve hides a lot of power behind a fairly compact interface. Before touching a single shortcut, get comfortable with four zones.
Media pool, source monitor, and timeline
The media pool holds your footage. The source monitor is where you preview and mark a clip before it ever touches your project. The timeline — the large area at the bottom — is the actual edit. The viewer above it shows either the source monitor or the timeline viewer, depending on which is active, and the inspector on the right changes based on what is selected.
New editors often skip the source monitor entirely, dragging straight from the media pool to the timeline. That works for simple projects, but it removes your ability to set precise in and out points, which is the foundation of efficient editing.
Track headers and their four controls
On the left edge of the timeline, each video and audio track has a header with several small controls you will use constantly:
- Lock prevents a track from being modified. Lock your music track and you will never accidentally nudge it while trimming dialogue.
- Mute silences the track during playback and export, without removing anything. This is the fastest way to check a cut with and without camera audio.
- Solo isolates a track so you hear only that one.
- Destination / track patching buttons on the far left of the timeline control where new material lands. If you add a clip and it appears on the wrong track, this is almost always the culprit.
The snapping and link-selection toggles
The toolbar above the timeline holds two toggles that matter more than they look. Snapping pulls clips, playhead, and edit points into alignment with whatever is nearby — useful 90 percent of the time, infuriating the rest. Link selection, when enabled, treats a video clip and its attached audio as a single unit, so selecting one selects both. Turning link selection off lets you grab only the video and move it without dragging its audio along. That one toggle prevents an enormous number of accidents.
Detaching audio from video: three approaches
When a clip has synchronized sound, Resolve keeps the picture and sound linked. There are three distinct ways to break that relationship, and they do different things.
Linked selection off (temporary)
With link selection disabled, clicking a clip selects only the video or only the audio portion. You can then move, trim, or delete one stream while the other stays put. The link itself is never destroyed — the moment you re-enable link selection, the two streams behave as a pair again. This is the right choice when you want to nudge a clip's picture half a second later while keeping its audio in place.
Unlink (permanent separation, same timeline object)
Selecting a linked clip and choosing to unlink removes the association between the two streams. They stay exactly where they are and continue playing together, but they are now independent entities on their own tracks. This is useful when you plan to trim them on completely different rhythms — for example, keeping dialogue continuous while cutting the picture into short reaction shots.
Detach audio (creates a separate audio clip)
Detach audio goes one step further: it splits the linked pair into a video clip on the video track and a new audio clip on the associated audio track. Some editors treat detach audio and unlink as identical. They are not. Detach audio generates a distinct audio item you can now move to an entirely different track, apply track-level effects to, or replace with a cleaner recording.
A practical rule: use link selection off for quick nudges, unlink for editing the two streams independently but in place, and detach audio when the sound is going somewhere else — a dedicated dialogue track, a different part of the timeline, or the trash.
Removing audio: delete, mute, disable, or replace
Once audio is detached, the question becomes what to do with it. There are four reasonable answers, and picking the wrong one is the source of most "why is my timeline broken" moments.
Delete it
Deleting an audio clip removes it from the timeline. If ripple behavior is active, everything to the right shifts left to close the gap. If ripple is off, a hole remains. For dialogue replacement on a locked edit, delete and leave the gap; for building a rough assembly, delete with ripple on so you are not constantly patching holes.
Mute it
Muting keeps the clip present but silent in playback and export. This is the safest option during a work in progress because nothing is lost — you can unmute at any time and compare. It is also ideal for a music track you are undecided about. Mute it, watch the cut, and decide.
Disable it
Disabling a clip (rather than deleting) takes it out of the render while reserving its space on the timeline. This is valuable when you know a section will eventually be filled — a voiceover you have not recorded yet, a b-roll clip still rendering — but you cannot afford to have the rest of the timeline shift around while you wait.
Replace it
Replacing is what most professional projects actually do. Camera audio is removed and substituted with a lavaliere recording, a boom take, or a clean studio pickup. The mechanic is simple: detach the camera audio, mute or delete it, then drag the replacement onto a free audio track and align it by waveform. Keeping the original camera audio on a muted track underneath costs you nothing and saves you if the replacement turns out to be unusable for a specific line.
Adding clips: insert, overwrite, append, fit to fill
The other half of timeline hygiene is getting new material in without collateral damage. Resolve gives you several edit types, and each has a distinct behavior.
Insert versus overwrite
An insert edit pushes everything downstream to the right to make room for the new clip. An overwrite edit drops the new clip on top of whatever occupies that range, replacing it. In the standard keyboard layout, insert maps to F9 and overwrite to F10, which tells you how central these two are considered to be.
Use insert when the new clips are additions to the story. Use overwrite when you are deliberately replacing a section — a revised take, a corrected graphic, a new ending. The most common beginner error is using overwrite without realizing it, then discovering that forty seconds of carefully trimmed b-roll has vanished under a single click.
Append to end
Append places selected material after the last clip on the timeline, automatically landing on the correct destination tracks. For rough assemblies and social cuts, append is the fastest way to build a sequence. It also avoids the trap of inserting in the middle of a finished section and shifting everything.
Place on top
Place on top drops a clip at the playhead on the nearest free track above the current one, without disturbing anything underneath. This is the goto for b-roll, title cards, cutaways, and reaction shots. When you are layering several visuals over one continuous interview, place on top keeps the audio bed untouched.
Fit to fill and three-point editing
For promotional edits, fit to fill is quietly one of the best tools in the application. You mark an in and out on the timeline to define a gap of a fixed duration, then mark in and out on the source clip. Resolve stretches or compresses the source — within reason — to fill the gap exactly. It lets you drop a shot into a music-driven section without hunting for a perfectly timed take.
More generally, three-point editing is the professional habit worth building. Set two marks on the source and one on the timeline (or vice versa) and the third is derived automatically. Editors who work this way cut faster and make fewer placement errors, because every edit has a defined target.
Layering and syncing new clips
Adding clips is easy; adding clips that stay in sync is the skill.
Control where they land with track patching
Before any edit, confirm the destination buttons. If you want the new clip's video on V2 and its audio on A3, enable exactly those targets and disable the rest. This single habit eliminates the classic problem of a clip appearing on top of your main footage and hiding it.
Align by waveform
When you replace production audio with a separate recorder, the fastest alignment method is visual: expand the audio tracks so the waveforms are tall, find a sharp transient — a hand clap, a door close, a plosive — and line up the peaks. Zoom in until each frame is wide. For interviews, a single well-aligned word is enough to verify the rest.
Sync by timecode and auto-sync
If your camera and audio recorder were jam-synced or share timecode, use the automatic sync tools and let Resolve align by timecode or by waveform analysis. For multicamera shoots, syncing all angles to a single reference track before you start cutting will save hours later.
Handle drift on long takes
Consumer cameras and separate recorders do not always run at exactly the same rate. On takes longer than a few minutes, audio can be sample accurate at the start and several frames out by the end. The fix is not one big nudge but periodic realignment: cut the replacement audio at natural pauses every few minutes and re-align each segment. Producers of long-form interviews do this as a matter of routine.
After the audio is gone: room tone, ambience, and voice
Silence is not neutral. Remove camera audio from a room scene and the result sounds like a vacuum, which is instantly distracting. Three layers fix it.
Room tone and ambience
Record thirty seconds of the empty room on every shoot day, or salvage a few seconds from a take where nobody speaks. Lay it under the whole scene at low level so cuts between takes stop sounding like dropouts. For exterior scenes, replace room tone with an appropriate ambience bed — street, wind, distant traffic.
Voice and dubbing workflows
When dialogue is unusable or a project needs narration, there are more options than ever. Human voiceover is still the quality benchmark for performance-driven work. Synthetic voice tools are genuinely useful for scratch tracks, temp narration, localization drafts, and explainer content where a consistent neutral read matters more than dramatic range. A common workflow is to generate a temp voice track, cut the whole video against it, and then decide whether to record a human performance over the same timing. Because the synthetic track already matches the final pacing, the human take slots in without recutting.
For multilingual versions, the same principle applies: generate draft narration per language, check that timing fits the picture, then hand it to native speakers for the final pass. Machine output is a scaffolding tool, not a substitute for a performance.
Levels and loudness targets
After removal and replacement, check overall loudness. Dialogue typically peaks somewhere in the range of minus twelve to minus six decibels full scale, with average level noticeably lower. For streaming and social delivery, an integrated loudness around minus fourteen LUFS with a true peak ceiling near minus one dBTP is a workable target; broadcast deliverables often specify a different standard, so always check the brief. A limiter on the master bus plus a loudness meter will catch the worst problems before delivery.
A complete short-form workflow, start to finish
Here is how the pieces fit together for a typical vertical social video.
- Build a rough assembly. Append your best takes to the end of the timeline. Ignore audio for now; you are only solving the story.
- Detach camera audio. Select everything, detach audio, and mute the resulting audio clips rather than deleting them. You now have a safety net.
- Tighten the picture. Turn off link selection when you need to nudge a shot without moving sound, and re-enable it when you want them locked together again.
- Add a music bed. Place it on a dedicated audio track, lock that track, and set a rough level so it supports rather than competes.
- Add b-roll with place on top. Layer cutaways on V2 and above. Verify track patching first so clips land where you expect.
- Add voiceover or captions. Align narration to the picture, then align the picture to the narration if the read is the anchor. Add captions for sound-off viewing.
- Restore ambience. A light room tone or atmosphere layer under the whole piece stops it sounding sterile.
- Mix and check loudness. Balance music against voice, run the loudness meter, and audition on phone speakers — that is where most viewers will hear it.
Common mistakes and how to fix them
The audio comes back when you move the clip
You moved a clip and its audio followed, or you deleted video and the sound stayed. Almost always this is link selection. Check the toggle before your next move, and if you have already broken the relationship, undo and redo the operation with the correct link state.
New clips overwrite your edit
You intended an insert but got an overwrite, or vice versa. Learn which edit type your toolbar buttons trigger and verify before clicking. When in doubt, undo immediately rather than trying to repair the timeline by hand.
New audio lands on the wrong track
Track patching. The destination buttons determine where new media goes, and they persist between edits. Glance at them before every drop.
Sync drifts over a long take
Frame rate or clock mismatch. Re-align in segments at natural pauses instead of trying to fix it with a single offset.
Snapping fights your placement
Disable snapping for micro-adjustments, then re-enable it for structural edits. Fighting the magnet is a waste of time.
Everything sounds thin after removing camera audio
You removed the room. Add ambience and room tone underneath, and check that your music bed is not masking dialogue.
FAQ
Should I delete camera audio or just mute it?
Mute during the edit, delete only when you are certain. Muted clips cost nothing and preserve the option to recover a line.
What is the difference between detach audio and unlink?
Unlink separates the video and audio streams of a single timeline clip so they can be edited independently in place. Detach audio creates a separate audio clip on the audio track, which you can move, reroute, or replace.
How do I move a video clip without moving its audio?
Turn off link selection, move the video only, then turn link selection back on.
Do I need to sync audio manually every time?
No. If the sources share timecode or waveform features are clear, use automatic sync. Manual alignment is the fallback for difficult material or short segments where automatic matching fails.
What is a reasonable loudness target for social video?
Roughly minus fourteen LUFS integrated with a true peak near minus one dBTP is a widely used working target. Always defer to the platform or client specification when one exists.
Can I add clips without shifting the rest of the timeline?
Yes — use overwrite or place on top instead of insert. Overwrite replaces the range under the playhead; place on top adds a new layer above the existing content without disturbing it.
How do I keep a music track from moving while I edit dialogue?
Lock the music track. Locked tracks are immune to accidental trims and shifts, and you can unlock them the moment you need to make a deliberate change.
Is it worth keeping a muted copy of the original audio?
For narrative and interview work, yes. Keep the original on a muted track until the project is locked. It costs a track and saves a reshoot.
