• Wondering which camera, gear, computer, or software to buy? Ask in our Gear Guide.

Audio work for the feature

Hello,

I have been shooting my second feature called 500 grams. But now I have a problem. I know how to edit, how to color correct but I do not know nothing about audio work at all.

I recorded the dialogue using a Rode Video Mic Pro connected to the camera, manual audio levels, +20db switch on in the mic and the audio itself is good. But what to do with it in the post? Should I for instace add foley sounds or not, should I mix a 5.1 audio track or just use stereo?

The editing program that I use is Adobe Premiere 5.5.
 
...the audio itself is good.

Don't take this the wrong, way, but I find this hard to believe. It may be the best that you've ever done, but I seriously doubt that it is anything approaching solid quality.

Should I for instace add foley sounds or not...

Okay, first off, it's Foley with a capital "F." It is named after Jack Foley who codified the process of adding human related sounds like footsteps, punches, cloth, etc. In recent decades the line between sound effects and Foley has become blurred. Also, technically, Foley is performed live to picture; if you are manually cutting in footsteps, etc. they are sound effects.

Audio post is an integral part of making a film. At the "Hollywood" level - and even at the indie level if you have the budget and knowledge - all the sound between lines of dialog is stripped out and replaced with room tone. You did record room tone, didn't you? Then Foley is performed, sound effects created (or pulled from a library), spotted and edited. Then the score and other music are spotted.

... should I mix a 5.1 audio track or just use stereo?

The choice is yours.

At the budgeted level, a 5.1 surround mix is done at specifically designed rerecording facilities by highly experienced people. BTW, surround sound is cool, but a very large percentage of the sound, mostly dialog and much of the Foley, is in the center channel. The surround channels are reserved for special effects.

You may find this thread of interest:

http://www.indietalk.com/showthread.php?t=45351


You may also want to search for other threads by AudioPostExpert (Greg) and myself.
 
Last edited:
You can do it!

Wont be all that great but go for it!


Yes, Foley is a must.
Do clothing pass
Do footsteps (you can download these pretty easily)
Do breathing pass
do everything else, keys, cars etc.
If you can go build your own sound library.

You can find a lot of sounds to fill the gaps at freesound.org or http://www.sounddogs.com/


Mix 5.1 not so much, but you could try 3.0 (L, R, C) be prepared to render a stereo only for the low end festivals, the bigger ones may require something else..
 
I am working with this already, so far I have exported the audio as wav file to Audacity program,

then used compressor effect and also I have balanced the different actors voice levels by using gain.

Then I have also added room tone from freesound to the clips and added audio effects like ambulance, paint brush sounds etc.

The biggest problem is to know what to use from original sounds and what to add. For me the audio works already when using headphones, but maybe I am mistaken.

For instance in this:

999241_497819696953183_784168320_n.jpg


Ok she is painting and there are no dialogue. I have added there sounds for brush (http://freesound.org/people/acrober/sounds/86114/) and room tone but that is it.

And this is the main issue for me, it is really hard to know what sounds to add.
 
Last edited:
For me the audio works already when using headphones, but maybe I am mistaken..

Why don't you post a short clip so we can listen and advise?

And this is the main issue for me, it is really hard to know what sounds to add.

The sounds which you may or may not need to add are suggested by your story, how you want to tell it, what emotional responses you want to elicit from those who are watching your film and when you want to elicit those responses. Everything depends on these questions, there are no absolute rules which always apply when it comes to sound design!

G
 
I ... used compressor effect and also I have balanced the different actors voice levels by using gain.

Using a compressor will increase the background noise unless used with great care. It should not be used as a short cut to level dialog volume.

FYI - compression and limiting are not effects, they are dynamic processors. Effects include reverbs and delays.

... I have also added room tone from FreeSound...

Again, I need to correct you on terminology and process. Room tone is the "silence" or ambient sound of the set on which the dialog was recorded. Ambience or background sounds are considered sound effects.

The biggest problem is to know what to use from original sounds and what to add.

That's always a judgment call. Sounds created or pulled from professional sound libraries are usually recorded better and are more "detailed" than what was recorded on the set.

For me the audio works already when using headphones, but maybe I am mistaken.

Two issues here. First, you are already used to the audio that was recorded on set, so there is a comfortable familiarity to it. Second, headphones are probably the worst medium to use to audition and edit sounds. Quality speakers are called for. There's a reason why sound editors like me and rerecording mixers like APE use speakers costing thousands in sonically treated rooms.

she is painting and there are no dialogue. I have added there sounds for brush and room tone but that is it.

With just a picture it's impossible to tell if the sounds you chose are working with the visuals.

Ambiences are usually constructed, The sound editor/designer starts with a basic BG (background) appropriate for the setting. Then details are added to create the mood/emotional effect desired. For instance, for a city scene I have dozens of versions of "empty street" which is nothing but the "empty air" of streets that I recorded in various locations. I will choose one and then add traffic sounds (busy or minimal), sirens (close or distant), foot traffic (if I see pedestrians), conversation (if the visual cues call for it), etc.

The sounds of the paint brush are usually recorded during Foley sessions, as cutting sound from a sound effects library almost never fully covers the needs of the scene. I would pick a brush and a surface and stroke by stroke mimic the actions on the screen. I may go through several combinations until I find a combination that satisfies the emotional content of the visuals.

The Foley props used do not have to be the actual props used by the actors/characters. I once did a project about an artist. Even though we saw a fine brush on the screen I ended up using a 2" paint brush on fine sandpaper to create the brush strokes.

The first and hardest lesson is to be an objective listener.
 
http://www.youtube.com/watch?v=Mi1cFjS3bhE&feature=youtu.be

Test image, hidden youtube link. Sorry in Finnish language, no subtitles.

-I used gain to level up the dialogue & gain to level down her walking sounds (heels).
-I added room tone
-I added computer noise
-I added car go by

One thing I notice right away is that I can be more brave with the sound effects, volume wise. I have edited them really low, around -15db.

Well thanks for the advice so far:)
 
Not terrible but not very good either. The most obvious faults (in reverse order) are:

3. While there are no obvious jumps in the room tone's volume or quality which is good, the overall level of the room tone is far too high.

2. There is little perspective to the sound. You change the shot a couple of times, a master shot then a couple of CUs but the sound does not follow the change of visual perspective. The footsteps have some perspective but the clothes rustling Foley is too present in the master shot for example, the computer key sounds do not change position or perspective relative to the picture and it sounds like a very quiet car is driving through the room. The car has to sound muffled and far more distant, a low pass filter would be a good idea. Also, the car-by cuts in and out too abruptly. Just as with your visuals, the sound too requires focus, perspective and DOF.

1. There is little/no sound design, except for the aforementioned car-by and room tone, there is no sound design. This results in the usual no budget film feel of not much happening and no pace or energy. It would appear to an audience that there's nothing obviously wrong but nevertheless, they would feel there was something not quite right (surreal/unnatural) about the scene and that it was undramatic and boring!

With sound design you literally have to think outside the box! In real life our field of vision is relatively limited and our field of focus is even more limited, not so with sound. We can hear 360deg around us at all times, without even moving our head. What we are focusing on visually combined with this 360deg sound is what gives us our perception of the world around us. We are not consciously aware of many/most of the individual sounds in this 360deg sound world, they just form an indistinct part of our overall perception of the location we are in. However, take all these indistinct individual sounds away and our perception instantly knows something is wrong. The box you have to think outside of is the frame of the picture and in your scene, the only concession you have made to this fact is one car-by, which on it's own sounds isolated, out of place and informs our perception mechanism of nothing.

So, the first question to ask yourself is: What is outside of the frame, where is this room we are seeing? At the moment, this room (and the people inside it) appears to be in outer space, light years from anything or anyone, and that is why the scene feels surreal/unnatural and is boring. You have to make your scene believable to even stand a chance of an audience finding it interesting! To create a believable scene, you have to create an entire, believable environment, not just a visual location or even a visual location with some dialogue and Foley. Not creating an entire believable environment is the most common and frequently the biggest mistake made by indie film makers!!!

So, once we know where this room is, we can start designing sound for it and using that sound not only to make the scene believable but also to inject pace/energy and to create some emotional response! For example: Are there other similar rooms in the building next to the one we are seeing, what sort of rooms are they and what is happening in those rooms? Where is the building our room is in (ie. what environment exists outside the window)? Maybe there are other shots/scenes in your film which visually establish the environment beyond our room, in which case we will have to create an already defined aural environment otherwise we can place this room in any environment we want. If this room is an office in a school for instance, think of all the appropriate school noises, kids running about, talking, shouting laughing or fighting, either in the corridors, outside or both, school bells or announcements, doors or lockers closing or slamming, if you think about it the list is almost endless. We have similar sets of sounds if the room is an office in an office building, hospital, police station or any other type of building. We also have a set of potential sounds for where the building is located. Maybe there is a train station nearby, traffic, an airport or maybe it's in the country. Furthermore, none of the potential sounds in our environment are constant, traffic gets louder or quieter, heavier or more sparse, closer and more distant, same with children' sounds, construction noise, etc. We can use this fact to create contrast, pace and interest when not much else is occurring in the scene. We can also create emotional implications, by using sounds of children fighting and screaming rather than playing and laughing. Distant police sirens, screaming construction machines or screeching train sounds create a different emotional response to pleasant bird song, consumer lawn mowers and private light aircraft. All these environmental sounds can be quiet and mostly quite distant and very distant but just because they are quiet and/or distant does not mean they are unimportant. Done well they create a believable environment, an emotional atmosphere and pace/energy, all of which are fundamentally important to film making!!

Make sure you read the thread Alcove posted a link to!

G
 
Last edited:
Thank you for the advice, I have learned many things about your reply, AudioPostExpert. For the scene I started to think what happends outside the apartment and since there is construction going on and quite much traffic I based lot of sounds to that.
 
Ok, now I did add a combination of sound effects to audacity, then exported them to wav. file and added to the premiere and rendered a clip.

The problem is that when I mix all them togeter, there were sounds of traffic, sounds of cars going by etc and then use that with the clip, it just goes downhill. This because when all the sounds are mixed and added to the clip, then volume dropped you cannot find any details from the background, it is just a noise. And if I raise the volume, that does not help either, then it is just more noise with somekind of effects between the noise.

How do I deal with this kind of problem? Should I just use my microphone and record the room tone, since I live in the same apartment it is quite easy. Or should I just add individual sounds for the background so they would be more easily regonized?
 
You're correct; when you put all of the sounds in at the same time you get a wash of noise.

Randy Thom and other noted sound designers and rerecording mixers adhere to the "Two and one half rule;" an audience can only comprehend two and one half sounds at the same time. This is a limitation placed on us by the artificial reproduction of sound by a speaker system. When we are in the real world we hear in a spherical manner - up, down, left right, front, back - and all manner of myriad permutations thereof simultaneously, and we can locate where the sounds come from. Our brains "edit" all of these hundreds, even thousands, of sounds down to those essentials required for our survival or cognitive needs. When reproduced by a speaker system sounds become very artificially focused, so when we have dozens of sounds happening simultaneously in a mix it becomes a confused mash. This is where the 2 1/2 rule applies.

This is not to say that there can only be three sounds present in the mix at any given time. For example, footsteps, cloth and breathing when combined in a mix are - in essence - the single combined sound of a person walking. In real life we would "tune out" traffic sounds until they are interrupted by something unusual; a honking horn, a squeal of brakes, a siren, etc. These are cues that inform us to focus on where the sounds are coming from and to be aware of potential dangers. But when general traffic sounds are reproduced they become a confused sonic mess. During the film experience we need visual cues to supply focus.

It is difficult to give specific advice without seeing the visuals so I'll create my own. We see the artist in a wide shot. There are traffic sounds outside. We hear a honking horn. Then we hear pedestrians. As the shot zooms in to the artist the traffic sounds fall back into the mix. The sound of the brush on canvas begins to become more dominant, and we start to hear the artist breathing. We eventually reach the same emotional point as the artist; all we hear is the brush on canvas, her breathing and perhaps some cloth Foley. She has tuned out the outside world in her creative concentration, and we sonically go along for the ride. The traffic is now an almost inaudible murmur in the background. Perhaps she is listening to music, or maybe the score is providing emotional content. Suddenly there is a squeal of brakes and a crash (we might mix it fairly loud for "shock" value) that pulls her out of her creative isolation; the traffic sounds suddenly become louder, sirens are heard in the distance, perhaps there are screams from onlookers. We have changed the point of view for the character and the audience.

The point is we constantly adjust the mix to adhere to (or contrast with) the characters point of view, or we focus on something else of which the character is not aware to heighten the tension. The sonic focus can change from minute to minute, even second by second. We pick what is important and subtract the irrelevancies.

This is not an easy task. Randy Thom (one of my personal heroes) talked about mixing for the Pixar animated film "Ratatouille." It took two full days to mix the 2.5 minute scene where Remy falls into and then tries to escape from the restaurant kitchen.

I'm sure that APE will chime in with a more detailed explanation.
 
I'm sure that APE will chime in with a more detailed explanation.

OK then. :)

I'm not a great fan of the 2½ rule, not because I believe it's entirely wrong but because it depends on what you mean by "comprehend" 2½ sounds at one time, how you define 2½ sounds and what you are trying to achieve with the mix at any one point in time. I tend to view it more as a consideration rather than as a rule but we've all got our different styles/methodologies.

Anyway, back to itarumaa...
The problem is that when I mix all them togeter, there were sounds of traffic, sounds of cars going by etc and then use that with the clip, it just goes downhill. This because when all the sounds are mixed and added to the clip, then volume dropped you cannot find any details from the background, it is just a noise. And if I raise the volume, that does not help either, then it is just more noise with somekind of effects between the noise.

How do I deal with this kind of problem? Should I just use my microphone and record the room tone, since I live in the same apartment it is quite easy. Or should I just add individual sounds for the background so they would be more easily regonized?

All I told you in my last post was a basic principle of sound design and the importance of creating a believable aural environment, I never said it was easy! In bigger budget commercial films they usually have a team of people editing these environments, then a SFX mixer to combine them and then a re-recording mixer who integrates them with the dialogue, Foley and music.

One of the difficulties of creating these environments is what Alcove alluded to, rather than wondering what sounds to add as per your first post, the problem is just as much; what sounds to leave out. In my previous post I explained how you could potentially have huge numbers of sounds in this environment but Alcove explained why you have to cut that number back at bit from what would likely exist in reality. You notice how I've always said "believable reality" or "believable environment", this is because what is believable when constructing an audio environment for film is not necessarily exactly the same as actual reality. This is an important point to consider but I'm not sure if it's the main cause of your current problem.

With any distant or background noise a component of the sound is noise and, noise is actually a quite complex and tricky thing to deal with because it behaves in slightly strange ways compared to most other types of sound. Unless you've successfully dealt with issue #3 I mentioned in my previous post then the problem you are now describing is most likely caused by far too much room tone (which is virtually all noise) interacting and combining with the noise in your traffic sound FX, resulting in just a wash of noise with no distinct sounds, just as you describe. Getting rid of enough room tone to avoid this problem, particularly the room tone recorded with your production dialogue is going to be very difficult and very probably impossible. This is why professional filmmakers always use mics on booms and never mic's mounted to the camera because mics mounted on the camera always record way too much room tone along with the dialogue and make creating a believable audio environment in post production very difficult or impossible. You are almost certainly going to have to live with a compromised audio environment and just do the best you can to remove as much noise (room tone) as possible from your production sound without ruining the audio quality of your dialogue. If I am correct and this is what's causing your problem then obviously recording more room tone in your apartment is only going to make the problem worse.

G
 
Last edited:
Interesting info about the room tone and booms etc.

I think it does make a alot of sense, since if I just add effects without the room tone, like for instance cars in the distance or keyboard sounds, I do like the results. Adding the room tone just makes the clip much worse.

Thanks again, this info will help to understand the audio work and will improve the quality of the sounds.
 
Back
Top