30-second answer: Advanced audio tools in Edits help you move beyond a single soundtrack: separate speech, music, and effects, balance their levels, clean up noise, and shape sound over time so the mix supports the picture. Start with clean recordings, give speech priority, keep music underneath it, use fades and ducking to make room, and check the mix on more than one device before export. Advanced tools do not rescue bad audio; they refine good audio.
On this page
- Think in layers, not one track
- Give speech priority in the mix
- Clean noise before adding polish
- Shape sound over time with fades
- Use ducking and balance deliberately
- Check the mix everywhere
- Using this in a desktop workflow
Most videos fail on sound before they fail on picture: viewers forgive a shaky shot far more readily than they forgive muffled speech, sudden volume jumps, or music that drowns the words. Edits runs on your phone, and its audio tools work within the app on the clips and layers in your project; this tutorial teaches the advanced discipline of layered, balanced sound, rather than promising a particular control, because controls change while the principles of layers, priority, and balance do not.
This is the last tutorial in the chain, and it builds on earlier ones: clean recordings from our voiceover tutorial, soundtracks from our music and audio tutorial, and clean exports from our export settings tutorial. If you have not read those, start there, because advanced tools applied to noisy, unbalanced source audio will only polish the problems. When you are done here, return to the Edits tutorials hub to revisit any topic.
Step 1: Think in layers, not one track
Think of your video’s sound as layers: speech on top, music beneath it, and effects — clicks, ambience, or emphasis sounds — underneath or between them. Each layer has a job, and each should be audible for a reason, at a level that suits that job. A mix built as layers is easy to adjust, because you can lower music without touching speech, or remove an effect without rebuilding the whole soundtrack.
Name and organise the layers in your head before you start adjusting: what is speech, what is music, what is effect, and what is ambience? If a sound does not fit a layer or has no job, it is a candidate for removal. Many muddy mixes are not caused by bad tools, but by too many sounds competing at once, and the first advanced step is simply to decide which sounds belong and which do not.
Step 2: Give speech priority in the mix
Give speech priority, always, unless the video is deliberately music-led. Set speech to a comfortable, consistent level first, then bring music and effects up underneath it until they support without competing, as in our voiceover tutorial. If a viewer must strain to hear words, the mix has failed, however good the music sounds on its own. Test intelligibility by listening with your eyes closed: if you can follow every word without effort, speech is at the right level.
Keep speech levels consistent across clips, because a viewer who must keep adjusting volume will stop watching. If one clip is much quieter than another, raise or lower it to match the rest before adding music, rather than expecting music to hide the difference. Consistency is more important than loudness: a mix that is uniformly a little quiet is better than one that jumps between loud and quiet.
Step 3: Clean noise before adding polish
Clean noise before adding polish, because effects and music layered over noise only make the noise more noticeable. Remove or reduce hum, hiss, wind, and room echo at the source where possible — by re-recording, by choosing a quieter take, or by trimming noisy sections — and use any noise-reduction tool lightly, because heavy reduction makes voices sound hollow or robotic. A light clean-up that preserves natural voice quality is better than an aggressive one that removes noise and voice together.
Check cleaned audio on headphones, where noise and artefacts are most audible, and compare it with the original: if the cleaned version sounds worse in any way, use less reduction or a different take. Noise that is only audible on headphones and not on speakers may be acceptable, but noise that is audible on speakers must be fixed. Never assume a viewer will listen on the same device you used to mix.
Step 4: Shape sound over time with fades
Shape sound over time with fades: fade music in at the start, fade it out at the end, and fade between sections so changes in level are gradual rather than abrupt, as in our transitions tutorial. A sudden jump from silence to full music, or from loud music to silence, startles viewers and feels amateur, while a short fade makes the same change feel deliberate. Fades are the audio equivalent of a smooth cut, and they cost almost nothing to apply.
Use fades on speech and effects as well, not just music: a voice that starts abruptly, or an effect that stops dead, draws attention to itself. Keep fades short — a second or two is usually enough — unless the moment calls for a long, gradual change, such as a scene ending. As with keyframes in our keyframes tutorial, small, purposeful changes over time sound more professional than large, sudden ones.
Step 5: Use ducking and balance deliberately
Use ducking — lowering music automatically or manually while speech plays — deliberately, so music supports speech instead of competing with it. When speech starts, music should drop to a level where words are clear; when speech ends, music can rise again. The drop should be smooth, not sudden, and the music’s resting level should be consistent, so viewers are not distracted by music constantly rising and falling for no reason.
Balance the whole mix by checking each layer’s contribution: can you hear speech clearly, music as a mood, and effects as emphasis, without any one layer dominating unintentionally? If an effect is louder than speech, it is too loud, however satisfying it sounds alone. Adjust levels in small steps, and re-check after each change, because a change to one layer affects how every other layer is perceived.
Step 6: Check the mix everywhere
Check the finished mix on more than one device — phone speaker, headphones, and, if possible, a larger speaker — because a mix that sounds balanced on one device can sound very different on another. Phone speakers emphasise midrange and lose bass, headphones reveal noise, and larger speakers reveal low-frequency problems. A mix that works on all three is ready; a mix that works on only one is not.
Check at different volumes as well, because a mix that is clear at full volume may be inaudible at low volume, and one that is comfortable at low volume may be harsh at full volume. Finally, export a test file and listen to it as a viewer would, away from the editing app, using our export settings tutorial. If anything sounds wrong in the exported file, fix it in the project and re-export, rather than assuming viewers will not notice.
Common advanced audio mistakes
The most common mistake is adding layers without priority, so speech, music, and effects all compete at full volume. The second is polishing noisy audio instead of cleaning or re-recording it, which preserves the problem under a layer of effects. The third is ignoring fades, so every sound change is abrupt.
Another mistake is mixing on one device only, then discovering the mix sounds wrong everywhere else. Finally, many creators never check the exported file, and publish a mix that sounded fine in the app but not as a finished video. The export check is the last, essential step, not an optional extra.
Using this in a desktop workflow
Advanced audio work benefits from a desktop setup, even though the edit happens in Edits on your phone. Use your computer to keep clean recordings, music, and effect files organised, with clear names and a note of each file’s level and purpose, so mixes can be rebuilt or revised without hunting for files. A larger screen helps with comparing layers, noting level changes, and keeping a mix log — what was raised, lowered, faded, or ducked, and why.
Keep a short mix checklist on your computer and run through it before every export: speech clear and consistent, music ducked under words, effects purposeful and not louder than speech, fades in place at the start and end, and the exported file checked on a phone speaker and headphones. A checklist takes a minute and catches the mistakes that are hardest to notice after hours of editing, when your ears have adjusted to a mix that a fresh viewer would hear as unbalanced. If you edit on an emulator, the larger display helps with precise level and fade adjustments, but always check the final mix on a phone speaker and headphones, since those are how most viewers will hear it. Keep a reference mix — a short video whose sound you admire — and compare your mix with it, to calibrate your ears. Good sound is a habit of layers, priority, fades, and checking, and that habit transfers to every video you make.
Frequently asked questions
What makes audio advanced?
Layered, prioritised sound: speech first, music beneath it, effects as emphasis, with fades, ducking, and balance checked on several devices.
Should speech always come first?
Yes, unless the video is deliberately music-led. If viewers must strain to hear words, the mix has failed.
How much noise reduction should I use?
As little as needed. Heavy reduction makes voices hollow; a light clean-up that preserves natural voice quality is better.
What is ducking?
Lowering music while speech plays, smoothly, so music supports words instead of competing with them.
Why check on several devices?
Because phone speakers, headphones, and larger speakers reveal different problems, and a mix must work on all of them.
Should I check the exported file?
Yes, always, as a viewer would, away from the editing app. Fix any problem in the project and re-export.
Back to the start: Edits Tutorials hub →
Download on Google Play Download on the App Store
Unofficial guide: this page is not published by Instagram or Meta. The store buttons above link to the official Google Play and App Store listings.