-
Wondering which camera, gear, computer, or software to buy? Ask in our Gear Guide.
Can turning up sound volume, destroy quality?
- Thread starter MelonDome
- Start date
Input volume is only part of the equation. The trick to getting good sound is where you put your mic.
If you are recording dialogue, don't record it from across the room, otherwise, turning up the input to record a weak voice signal will result in more ambient noise. The trick is to get the mic as close to your subject as possible. You should turn your camera mic's input high enough that you are getting a strong signal from your subject.
I usually use sound from my closeup takes, where the mic is about a foot or two from the actor's mouth. Even if I go to a long shot, I can always EQ that same sound so it seems natural (getting rid of the "proximity" effect).
I always recommend a mic that isn't just built into the camera. Get yourself a Sennheiser ME66 or similar. Either use a boom pole or if you don't have help, use a mic stand with a boom arm.
Another recommendation is to record 2 level settings. Some cameras like the DVX100 or HVX200 will allow you to record 2 levels and only use 1 microphone. AWESOME FEATURE! To do this, set both Inputs to INPUT 2. Set one input to optimal (just under redline) and set the second input much lower. That way, if the actor shouts way into the red zone, you can use your backup track that isn't distorted, because it is set at a lower level. Your editing program can split these different leveled left and right tracks, ie replace the Right track with the audio from Left and vice versa.
Alternately, if you can't afford a decent mic, record as normal, so you have a source track that your actors can hear. Have them come over and rerecord their lines (ADR). The hard part is for an actor to recreate his/her energy from a room, so that source track is critical for him/her to hear and compare to. Many scenes (dialogue in a convertible, while driving, a fast boat, loud industrial noises in the background) will require you to dub, regardless. Trying to filter/fix the sound can only do so much, but ADR is clean. Always record room tone or background noise for a good 30 - 60 seconds. You can later use it on a separate track, but with total control of the volume.
If you don't have a sound booth (I built one with hinges, wood and bed foam), use a closet with carpet and lots of clothes, for absorbtion, so you don't get reflections or room echo. You can always add slight room reverb in post, if needed, and separate ambient (background noise) tracks. Do this and your projects will sound super!
If you have a low level track and no other option, try using a compressor to bring it up. I had an ultra quiet track, with the actor's only line and was able to save it this way. Fortunately, the set was very quiet, so the noise floor wasn't much of a factor.
If you are recording dialogue, don't record it from across the room, otherwise, turning up the input to record a weak voice signal will result in more ambient noise. The trick is to get the mic as close to your subject as possible. You should turn your camera mic's input high enough that you are getting a strong signal from your subject.
I usually use sound from my closeup takes, where the mic is about a foot or two from the actor's mouth. Even if I go to a long shot, I can always EQ that same sound so it seems natural (getting rid of the "proximity" effect).
I always recommend a mic that isn't just built into the camera. Get yourself a Sennheiser ME66 or similar. Either use a boom pole or if you don't have help, use a mic stand with a boom arm.
Another recommendation is to record 2 level settings. Some cameras like the DVX100 or HVX200 will allow you to record 2 levels and only use 1 microphone. AWESOME FEATURE! To do this, set both Inputs to INPUT 2. Set one input to optimal (just under redline) and set the second input much lower. That way, if the actor shouts way into the red zone, you can use your backup track that isn't distorted, because it is set at a lower level. Your editing program can split these different leveled left and right tracks, ie replace the Right track with the audio from Left and vice versa.
Alternately, if you can't afford a decent mic, record as normal, so you have a source track that your actors can hear. Have them come over and rerecord their lines (ADR). The hard part is for an actor to recreate his/her energy from a room, so that source track is critical for him/her to hear and compare to. Many scenes (dialogue in a convertible, while driving, a fast boat, loud industrial noises in the background) will require you to dub, regardless. Trying to filter/fix the sound can only do so much, but ADR is clean. Always record room tone or background noise for a good 30 - 60 seconds. You can later use it on a separate track, but with total control of the volume.
If you don't have a sound booth (I built one with hinges, wood and bed foam), use a closet with carpet and lots of clothes, for absorbtion, so you don't get reflections or room echo. You can always add slight room reverb in post, if needed, and separate ambient (background noise) tracks. Do this and your projects will sound super!
If you have a low level track and no other option, try using a compressor to bring it up. I had an ultra quiet track, with the actor's only line and was able to save it this way. Fortunately, the set was very quiet, so the noise floor wasn't much of a factor.
Last edited:
In the digital age (I started out when analog tape hiss was sometimes a major issue) low volume levels are not as much of a problem as they used to be.
The noise problems with production sound are usually the ambient sound - wind, traffic, etc. - or hiss from budget sound gear; cheap mics, pre-amps and the like. Even the most advanced cameras have minimal audio implementation and substandard (by an audio pros standards) audio hardware.
As Scoopic mentioned capturing quality production sound is job one, something most indie filmmakers tend to ignore until it is much too late. Most of my work is attempting to make horrible production sound listenable, i.e. you can now understand what the actors are saying but not much else.
The equipment noise problems can be mitigated by proper gain-staging which is properly setting the levels between the various pieces of audio gear. If done properly this may also increase the volume levels into your camera.
All that being said, low signal levels into the camera - if you have paid attention to the ambient and equipment noise problems - should not be much of a problem. There are audio plug-ins that can increase the volume levels of your audio tracks, most commonly called "Gain" or "Normalize", and are included in most NLEs.
You open the plug-in, select the audio clip and (in Pro Tools, I don't know what it would be in Final Cut) select "Find Peak". The plug-in will analyze the clip and find the highest dB (decibel) level in the clip. If the plug-in shows -12 as the highest dB level you can increase the gain by up to 12dB, although I would usually leave about 2dB of extra headroom (increasing the dB level to -2dB). There are two things to keep in mind. The first is that the volume of everything will be increased - the noise, footsteps, whatever - as well as the dialog. The second is that it only shows the highest peak. So if you have two characters whispering to each other (-34dB) and then there is a gunshot (-4dB) the peak for the entire clip is -4dB. A 2dB increase will barely be noticeable. What you do in this case is cut out the gunshot and place it on a separate audio track; then run the gain/normalize plug-in on the remaining whispered dialog. You can now increase that dialog by up to 34dB, although as previously mentioned all of the other sounds will increase in volume as well.
You can apply gain/normalize to the entire mix if needed.
I hope that this hasn't been overly confusing and was of some help.
The noise problems with production sound are usually the ambient sound - wind, traffic, etc. - or hiss from budget sound gear; cheap mics, pre-amps and the like. Even the most advanced cameras have minimal audio implementation and substandard (by an audio pros standards) audio hardware.
As Scoopic mentioned capturing quality production sound is job one, something most indie filmmakers tend to ignore until it is much too late. Most of my work is attempting to make horrible production sound listenable, i.e. you can now understand what the actors are saying but not much else.
The equipment noise problems can be mitigated by proper gain-staging which is properly setting the levels between the various pieces of audio gear. If done properly this may also increase the volume levels into your camera.
All that being said, low signal levels into the camera - if you have paid attention to the ambient and equipment noise problems - should not be much of a problem. There are audio plug-ins that can increase the volume levels of your audio tracks, most commonly called "Gain" or "Normalize", and are included in most NLEs.
You open the plug-in, select the audio clip and (in Pro Tools, I don't know what it would be in Final Cut) select "Find Peak". The plug-in will analyze the clip and find the highest dB (decibel) level in the clip. If the plug-in shows -12 as the highest dB level you can increase the gain by up to 12dB, although I would usually leave about 2dB of extra headroom (increasing the dB level to -2dB). There are two things to keep in mind. The first is that the volume of everything will be increased - the noise, footsteps, whatever - as well as the dialog. The second is that it only shows the highest peak. So if you have two characters whispering to each other (-34dB) and then there is a gunshot (-4dB) the peak for the entire clip is -4dB. A 2dB increase will barely be noticeable. What you do in this case is cut out the gunshot and place it on a separate audio track; then run the gain/normalize plug-in on the remaining whispered dialog. You can now increase that dialog by up to 34dB, although as previously mentioned all of the other sounds will increase in volume as well.
You can apply gain/normalize to the entire mix if needed.
I hope that this hasn't been overly confusing and was of some help.
To answer your actual question - it depends. As I mentioned in my previous post a lot of it is your gain structure. The whole key is to avoid distortion, which is pretty easy - don't let any audio tracks get into the red in your mix. As Scoopic mentioned a compressor can be used to increase volume, but using compression incorrectly can actually INCREASE the background noise on your audio tracks or can cause "pumping" or "breathing" which can sound cool when used on some music tracks but is a disaster for film sound. Compression can be used to increase "apparent" loudness without a real increase in volume; it sounds louder but the dB levels are not much louder. I'm sure that you've noticed that affect in commercials. You can also increase "presence" with judicious use of EQ (equalization). Conversely, the misuse of EQ can have a major negative effect on your audio, making it sound "honky" or "tinny", etc.
I'll try and put together a blog on some audio and audio post basics.