{"course":{"slug":"video-editing","title":"Editing Video","url":"https://addaly.com/learn/video-editing","level":"beginner","track":"create"},"lesson":{"slug":"sound-is-most-of-it","title":"Bad sound kills a good picture","url":"https://addaly.com/learn/video-editing/sound-is-most-of-it","position":30,"of":77,"minutes":11,"answer":"Get the mic within 30 cm, keep peaks near -6 dBFS, record thirty seconds of room tone, and use less noise reduction than you want to.","keyIdea":"Get the mic within 30 cm, keep peaks near -6 dBFS, record thirty seconds of room tone, and use less noise reduction than you want to."},"body":{"format":"markdown","text":"## The asymmetry nobody warns you about\n\nA soft, badly lit, wobbly picture with clean sound reads as low-budget but competent. People watch it to the end. A sharp, well-composed picture with hissy, echoing, uneven sound reads as broken, and people leave in the first fifteen seconds.\n\nThis is not taste. Speech intelligibility is effortful: if a listener has to reconstruct words from a noisy signal, they spend attention on decoding rather than on what you are saying, and after twenty seconds of that they stop. Nothing in the picture compensates, because nothing in the picture is doing that work. If time on a project is short, spend it on the audio.\n\n## The numbers that actually matter\n\nDigital audio is measured in **dBFS** (decibels relative to full scale) counting down from zero. Zero is a hard ceiling, not a warning: above it, samples are flattened, the waveform develops a square edge, and it crackles. Unlike analogue tape there is no warmth up there and nothing to recover. Clipping is permanent damage.\n\nWorking targets for spoken word:\n\n- Dialogue **peaks** around **-6 dBFS**, average level around **-12 dBFS**. That leaves headroom for a laugh or a raised voice.\n- Music sitting under dialogue: roughly **-18 to -24 dBFS**, quieter than feels right to you, because you already know the words.\n- The final mix is normalised for the platform at export.\n\nThe two commonest phone-video faults are opposite versions of one mistake: gain at maximum so everything clips, or recording so quietly that you add 20 dB in post, which raises the hiss and the traffic and the fridge by 20 dB too. The noise floor is baked in at the moment of recording.\n\n## Distance beats equipment\n\nIf you buy one thing, buy a cheap wired lavalier microphone (around ₹800 to ₹1,500, or ten to twenty dollars) and clip it about 20 to 30 centimetres from the mouth, out of frame. It will beat an expensive microphone two metres away, every time.\n\nThe reason is the ratio of direct sound to reflected sound. Sound reaching the mic straight from the mouth falls off quickly with distance; sound bouncing off walls and ceiling is spread evenly through the room and barely falls off at all. Halve the distance and you dramatically improve that ratio. This is why phone video shot across a room sounds like a bathroom: not because the phone mic is bad, but because most of what reaches it has been off a wall first.\n\nThen turn off the fan, the air conditioner and the fridge, close the window, and put a blanket behind the camera if the room is bare. None of it costs anything.\n\n## Room tone, and the holes in your dialogue\n\nRecord **thirty seconds of the room doing nothing** at every location. Everyone forgets, and everyone regrets it.\n\nEvery room has a noise floor: air, hum, distant traffic. When you cut dialogue, that floor stops and restarts at every join, and the ear hears the sudden silence as a hole punched in the scene. It is one of the clearest amateur tells there is. Lay room tone on a track under the whole scene, fill any true gaps with it, and the holes vanish.\n\nWhile you are there: a **high-pass filter at 80 to 100 Hz** on every voice track removes rumble, handling noise and air conditioning without touching the speech.\n\n> Figure: Four things that cost nothing and fix most bad sound. Steps: 1. Microphone within 30 cm (A cheap wired lavalier close beats an expensive microphone across the room); 2. Peaks around -6 dBFS (Recording quietly bakes the hiss in; going over zero is permanent damage); 3. Thirty seconds of room tone (Every location, every time. It fills the holes your cuts punch in the background); 4. High-pass at 80 to 100 Hz (Removes rumble, footsteps and air conditioning, and touches no speech). Only the fourth happens in the edit. The first three cannot be added afterwards, which is why the noise floor is decided at the moment of recording.\n\n## Noise reduction, and exactly how it fails\n\nClassic noise reduction works by spectral subtraction. You give it a sample of noise alone, it builds a profile of how much energy sits in each frequency band, and it subtracts that amount everywhere.\n\nPush it too far and quiet harmonics of the voice fall below the estimate, so they are removed in some frames and kept in others. Those fragments flutter in and out and you get the watery, swirling, underwater artefact engineers call musical noise.\n\nThe rule: **if you can hear the noise reduction, there is too much of it.** Six decibels that leaves a little honest hiss beats twenty that leaves a robot. Audiences forgive background noise. They do not forgive artefacts, because artefacts sound like something is wrong with the file.\n\nFree tools: Audacity's noise reduction, and the EQ and dynamics on Resolve's Fairlight page, though some Resolve restoration plug-ins are Studio-only.\n\n## The newer speech tools, and where the line is\n\nA different class of tool does not subtract noise. It runs the recording through a model and **resynthesises the voice**. The results can be startling: a phone recording from across a room comes back close to studio.\n\nTwo cautions. It is generating audio, so it can shift timbre and invent detail that was never there, which makes it inappropriate wherever the recording is evidence, journalism, or a record of what someone said. And once a tool goes past cleanup into replacing words, changing accents or producing a voice, that is synthetic media: it needs a visible label (several countries, India among them, now require labelling of synthetic content), and somebody's voice needs their written consent. Consent for a likeness or a voice is not a style question. [Making Things with AI](/learn/making-things-with-ai) covers where that stands.\n\n## Two traps that catch everyone once\n\n- **Mono recorded into one channel.** A lavalier plugged into a single input often lands on the left channel only, so anyone listening on one earbud hears silence. Set the clip's audio to mono, or copy it to both channels.\n- **No sync reference.** If picture and sound came from different devices, clap once, hard, in frame at the top of every take. A clap gives you a single unmistakable spike to line up. Waveform auto-sync works well but fails on noisy or quiet takes, and then you will want that clap.\n\n## Today\n\nPlay something you made with the picture turned off. If you would not listen to it as a podcast, the picture is not saving it."},"check":{"question":"An interview recorded near a busy road is cleaned up with heavy noise reduction. The traffic is gone, but the voice now has a swirling, underwater quality. What is happening?","options":[{"text":"The high-pass filter was set too high and has removed the fundamental frequency of the voice, leaving only harmonics.","correct":false,"tempting":"Blames a filter that cuts a fixed low band, which produces thinness rather than fluctuating artefacts."},{"text":"The reduction is subtracting a noise estimate from every band, and quiet voice harmonics near that threshold are being removed in some frames and kept in others, so they flicker.","correct":true},{"text":"The recording was clipped at source, and noise reduction has exposed the distortion that the traffic was previously masking.","correct":false,"tempting":"Confuses permanent clipping damage, which sounds like crackle, with an artefact introduced by the processing itself."},{"text":"The noise profile was sampled from a section containing speech, so the tool is reducing the whole signal evenly.","correct":false,"tempting":"Identifies a real setup error, but an over-inclusive profile dulls the voice consistently rather than making it fluctuate."}]},"prev":{"slug":"archive-and-other-peoples-footage","title":"Using material you did not shoot","url":"https://addaly.com/learn/video-editing/archive-and-other-peoples-footage"},"next":{"slug":"music-and-the-claim","title":"The music you found is not free","url":"https://addaly.com/learn/video-editing/music-and-the-claim"},"authorship":"Founder and series editor: Dr Mustafa Mun. Written by AI models to a written standard. Exam keys checked blind by a second model: 3,665 of 3,666 agreed. Academic review invited.","review":{"status":"Draft, open for academic review"},"cite":"Addaly (2026) 'Bad sound kills a good picture', in Editing Video. Edited by M. Mun. Addaly. Available at: https://addaly.com/learn/video-editing/sound-is-most-of-it (Accessed: 31 October 2026).","licence":{"name":"CC BY-NC-SA 4.0","url":"https://creativecommons.org/licenses/by-nc-sa/4.0/"},"reader":{"schema":1,"title":"Bad sound kills a good picture","courseTitle":"Editing Video","module":{"index":4,"slug":"sound","title":"Sound, which is most of it"},"answer":null,"keyIdea":"Get the mic within 30 cm, keep peaks near -6 dBFS, record thirty seconds of room tone, and use less noise reduction than you want to.","hero":{"kind":"figure","figure":{"t":"figure","label":"Figure 4.1","figure":{"type":"flow","steps":[{"label":"Microphone within 30 cm","detail":"A cheap wired lavalier close beats an expensive microphone across the room"},{"label":"Peaks around -6 dBFS","detail":"Recording quietly bakes the hiss in; going over zero is permanent damage"},{"label":"Thirty seconds of room tone","detail":"Every location, every time. It fills the holes your cuts punch in the background"},{"label":"High-pass at 80 to 100 Hz","detail":"Removes rumble, footsteps and air conditioning, and touches no speech"}],"title":"Four things that cost nothing and fix most bad sound","caption":"Only the fourth happens in the edit. The first three cannot be added afterwards, which is why the noise floor is decided at the moment of recording."},"alt":"Figure: Four things that cost nothing and fix most bad sound. Steps: 1. Microphone within 30 cm (A cheap wired lavalier close beats an expensive microphone across the room); 2. Peaks around -6 dBFS (Recording quietly bakes the hiss in; going over zero is permanent damage); 3. Thirty seconds of room tone (Every location, every time. It fills the holes your cuts punch in the background); 4. High-pass at 80 to 100 Hz (Removes rumble, footsteps and air conditioning, and touches no speech). Only the fourth happens in the edit. The first three cannot be added afterwards, which is why the noise floor is decided at the moment of recording."}},"blocks":[{"t":"h2","id":"the-asymmetry-nobody-warns-you-about","runs":[{"text":"The asymmetry nobody warns you about"}]},{"t":"p","runs":[{"text":"A soft, badly lit, wobbly picture with clean sound reads as low-budget but competent. People watch it to the end. A sharp, well-composed picture with hissy, echoing, uneven sound reads as broken, and people leave in the first fifteen seconds."}]},{"t":"p","runs":[{"text":"This is not taste. Speech intelligibility is effortful: if a listener has to reconstruct words from a noisy signal, they spend attention on decoding rather than on what you are saying, and after twenty seconds of that they stop. Nothing in the picture compensates, because nothing in the picture is doing that work. If time on a project is short, spend it on the audio."}]},{"t":"h2","id":"the-numbers-that-actually-matter","runs":[{"text":"The numbers that actually matter"}]},{"t":"p","runs":[{"text":"Digital audio is measured in "},{"text":"dBFS","b":true},{"text":" (decibels relative to full scale) counting down from zero. Zero is a hard ceiling, not a warning: above it, samples are flattened, the waveform develops a square edge, and it crackles. Unlike analogue tape there is no warmth up there and nothing to recover. Clipping is permanent damage."}]},{"t":"p","runs":[{"text":"Working targets for spoken word:"}]},{"t":"ul","items":[[{"text":"Dialogue "},{"text":"peaks","b":true},{"text":" around "},{"text":"-6 dBFS","b":true},{"text":", average level around "},{"text":"-12 dBFS","b":true},{"text":". That leaves headroom for a laugh or a raised voice."}],[{"text":"Music sitting under dialogue: roughly "},{"text":"-18 to -24 dBFS","b":true},{"text":", quieter than feels right to you, because you already know the words."}],[{"text":"The final mix is normalised for the platform at export."}]]},{"t":"p","runs":[{"text":"The two commonest phone-video faults are opposite versions of one mistake: gain at maximum so everything clips, or recording so quietly that you add 20 dB in post, which raises the hiss and the traffic and the fridge by 20 dB too. The noise floor is baked in at the moment of recording."}]},{"t":"h2","id":"distance-beats-equipment","runs":[{"text":"Distance beats equipment"}]},{"t":"p","runs":[{"text":"If you buy one thing, buy a cheap wired lavalier microphone (around ₹800 to ₹1,500, or ten to twenty dollars) and clip it about 20 to 30 centimetres from the mouth, out of frame. It will beat an expensive microphone two metres away, every time."}]},{"t":"p","runs":[{"text":"The reason is the ratio of direct sound to reflected sound. Sound reaching the mic straight from the mouth falls off quickly with distance; sound bouncing off walls and ceiling is spread evenly through the room and barely falls off at all. Halve the distance and you dramatically improve that ratio. This is why phone video shot across a room sounds like a bathroom: not because the phone mic is bad, but because most of what reaches it has been off a wall first."}]},{"t":"p","runs":[{"text":"Then turn off the fan, the air conditioner and the fridge, close the window, and put a blanket behind the camera if the room is bare. None of it costs anything."}]},{"t":"h2","id":"room-tone-and-the-holes-in-your-dialogue","runs":[{"text":"Room tone, and the holes in your dialogue"}]},{"t":"p","runs":[{"text":"Record "},{"text":"thirty seconds of the room doing nothing","b":true},{"text":" at every location. Everyone forgets, and everyone regrets it."}]},{"t":"p","runs":[{"text":"Every room has a noise floor: air, hum, distant traffic. When you cut dialogue, that floor stops and restarts at every join, and the ear hears the sudden silence as a hole punched in the scene. It is one of the clearest amateur tells there is. Lay room tone on a track under the whole scene, fill any true gaps with it, and the holes vanish."}]},{"t":"p","runs":[{"text":"While you are there: a "},{"text":"high-pass filter at 80 to 100 Hz","b":true},{"text":" on every voice track removes rumble, handling noise and air conditioning without touching the speech."}]},{"t":"h2","id":"noise-reduction-and-exactly-how-it-fails","runs":[{"text":"Noise reduction, and exactly how it fails"}]},{"t":"p","runs":[{"text":"Classic noise reduction works by spectral subtraction. You give it a sample of noise alone, it builds a profile of how much energy sits in each frequency band, and it subtracts that amount everywhere."}]},{"t":"p","runs":[{"text":"Push it too far and quiet harmonics of the voice fall below the estimate, so they are removed in some frames and kept in others. Those fragments flutter in and out and you get the watery, swirling, underwater artefact engineers call musical noise."}]},{"t":"p","runs":[{"text":"The rule: "},{"text":"if you can hear the noise reduction, there is too much of it.","b":true},{"text":" Six decibels that leaves a little honest hiss beats twenty that leaves a robot. Audiences forgive background noise. They do not forgive artefacts, because artefacts sound like something is wrong with the file."}]},{"t":"p","runs":[{"text":"Free tools: Audacity’s noise reduction, and the EQ and dynamics on Resolve’s Fairlight page, though some Resolve restoration plug-ins are Studio-only."}]},{"t":"h2","id":"the-newer-speech-tools-and-where-the-line-is","runs":[{"text":"The newer speech tools, and where the line is"}]},{"t":"p","runs":[{"text":"A different class of tool does not subtract noise. It runs the recording through a model and "},{"text":"resynthesises the voice","b":true},{"text":". The results can be startling: a phone recording from across a room comes back close to studio."}]},{"t":"p","runs":[{"text":"Two cautions. It is generating audio, so it can shift timbre and invent detail that was never there, which makes it inappropriate wherever the recording is evidence, journalism, or a record of what someone said. And once a tool goes past cleanup into replacing words, changing accents or producing a voice, that is synthetic media: it needs a visible label (several countries, India among them, now require labelling of synthetic content), and somebody’s voice needs their written consent. Consent for a likeness or a voice is not a style question. "},{"text":"Making Things with AI","href":"https://addaly.com/learn/making-things-with-ai"},{"text":" covers where that stands."}]},{"t":"h2","id":"two-traps-that-catch-everyone-once","runs":[{"text":"Two traps that catch everyone once"}]},{"t":"ul","items":[[{"text":"Mono recorded into one channel.","b":true},{"text":" A lavalier plugged into a single input often lands on the left channel only, so anyone listening on one earbud hears silence. Set the clip’s audio to mono, or copy it to both channels."}],[{"text":"No sync reference.","b":true},{"text":" If picture and sound came from different devices, clap once, hard, in frame at the top of every take. A clap gives you a single unmistakable spike to line up. Waveform auto-sync works well but fails on noisy or quiet takes, and then you will want that clap."}]]},{"t":"h2","id":"today","runs":[{"text":"Today"}]},{"t":"p","runs":[{"text":"Play something you made with the picture turned off. If you would not listen to it as a podcast, the picture is not saving it."}]}],"check":{"question":[{"text":"An interview recorded near a busy road is cleaned up with heavy noise reduction. The traffic is gone, but the voice now has a swirling, underwater quality. What is happening?"}],"options":[{"text":[{"text":"The high-pass filter was set too high and has removed the fundamental frequency of the voice, leaving only harmonics."}],"correct":false,"tempting":[{"text":"Blames a filter that cuts a fixed low band, which produces thinness rather than fluctuating artefacts."}]},{"text":[{"text":"The reduction is subtracting a noise estimate from every band, and quiet voice harmonics near that threshold are being removed in some frames and kept in others, so they flicker."}],"correct":true,"tempting":null},{"text":[{"text":"The recording was clipped at source, and noise reduction has exposed the distortion that the traffic was previously masking."}],"correct":false,"tempting":[{"text":"Confuses permanent clipping damage, which sounds like crackle, with an artefact introduced by the processing itself."}]},{"text":[{"text":"The noise profile was sampled from a section containing speech, so the tool is reducing the whole signal evenly."}],"correct":false,"tempting":[{"text":"Identifies a real setup error, but an over-inclusive profile dulls the voice consistently rather than making it fluctuate."}]}],"order":[0,2,1,3],"why":null},"prev":{"slug":"archive-and-other-peoples-footage","title":"Using material you did not shoot","minutes":10},"next":{"slug":"music-and-the-claim","title":"The music you found is not free","minutes":9}}}