The Principles of Sound Design

I posted a homework opportunity and continued conversation.

http://www.indietalk.com/showthread.php?p=306060#post306060

The above suggestion is an excellent exercise to achieve one particular goal, to get students to start thinking in terms of telling a story with sound rather than with visuals. In other respects though it's a poor exercise because it misses the point of sound design, which is how the sound and picture interact.

Click to expand...

The point was not just about sound, but how to pare down dialog. The radio play "assignment" is obvious, telling a story with sound alone. The silent film "assignment" is that when you need to rely on sound and picture ONLY - no dialog - you really need to think about what they can do that replaces dialog.

Alcove Audio said:

The silent film "assignment" is that when you need to rely on sound and picture ONLY - no dialog - you really need to think about what they can do that replaces dialog.

Click to expand...

Yep, sorry I didn't read carefully enough. When I read "silent film" my brain goes straight to picture and music, with no other sound. Even though I know sound was sometimes performed during screenings, the filmmakers had no direct input or control over these sound performances. The exercises I think are much more valid, now I understand them better but still what I said in my previous post is valid because it's the interaction of picture, sound and dialogue which is at the heart of sound design.

I once met a famous indian tabla guru. For those who don't know, tabla is a traditional indian instrument comprised of two drums tuned slightly differently, the left drum is played by the left hand and the right drum played by the right hand. The technique of tabla playing is extremely sophisticated and difficult. This guru told me his teaching technique, which stuck in my mind. His students study with him for 4 years. The first year they work only with the left hand, the second year they work only on the right hand, the third year they work on both hands together and the final year he teaches them to forget they have hands! What he meant was that after mastering the technique of tabla playing, in the final year he taught them about musicality and that technique was completely irrelevant except in terms of providing the physical ability to express their musicality.

G

AudioPostExpert said:

When I read "silent film" my brain goes straight to picture and music, with no other sound. Even though I know sound was sometimes performed during screenings, the filmmakers had no direct input or control over these sound performances.

Click to expand...

Back in the hey day of silent film the big movie houses had, in addition to the full orchestra, up to 20 people behind the screen supplying sound effects and even "vocalizations." (APE already knows this, so this is an FYI for everyone else.)

I once talked with a guy at TCM (Turner Classic Movies) about putting together a live sound effects team to do sound effects for the restored silents that they screened with a new score (also live) in conjunction with their Young Composers competition. They nixed the idea, of course, but it would have been great fun!


Just because I love this picture, the older sister of my paternal grandfather, Emily, did sound effects for silent film on her "traps" kit, the precursor to the modern drum kit. The more toys you had, the more gigs you got!

We may now have finally arrived at a much more precise definition of "Sound Design". Sound design can be defined as all the sonic properties of the soundtrack which are designed to lie beyond the boundary of reality, for the purpose of storytelling (audience manipulation). Good sound design therefore always exists between the boundary of reality and the boundary of believability. This definition is more accurate than our previous definition because it indicates the difference between sound design and sound editing/mixing. It is this definition of sound design which leads me to conclude that the vast majority of no/low budget films contain either little or absolutely no sound design, because the filmmakers are attempting (with varying levels of success) to recreate a sonic reality. Even with great skill and resources, the very best which can be achieved with this approach to sound is a finished film where the sound does not reduce the audience's perception of the visuals. In contrast, sound design exists solely to greatly enhance the audience's perception of the visuals!

Click to expand...

This thread is amazing. Completely makes sense, all of it. Thank you for taking the time to write it, A.P.E.

Love this sentence:

"Good sound design therefore always exists between the boundary of reality and the boundary of believability."

Indeed I think good sound design often might try to nudge itself as close as possible to the boundary of believability, while stretching the boundary of reality as far as it needs to go to cover the distance!

sylents said:

Indeed I think good sound design often might try to nudge itself as close as possible to the boundary of believability, while stretching the boundary of reality as far as it needs to go to cover the distance!

Click to expand...

In practice, once we get to post production, the boundary of reality is pretty much immovable. In reality if we close a certain door, in a certain way, in a certain room, it will always sound the same (with very minor variations). In audio post we can use the fact that there are always minor variations and make a change beyond a minor variation and still stay within the bounds of believability. This fact raises 3 points:

1. We can change where the actual reality point exists by having our actor close the door differently during filming, say slamming it instead of just closing it.

2. There are obviously limits to how far we can push believability. If an actor closes a door normally but we use the sound of a door being slammed very hard, we are probably going beyond limits of believability. Instead of enhancing the dramatic impact of the door closing, the audience will feel something is wrong (without necessarily knowing why) and we will have actually created the opposite effect desired and pulled them out of the scene rather than involving them more in it.

3. Point #2 above raises the very difficult issue of what is the limit of believability. It's particularly difficult for the lo/no budget filmmaker for two reasons: A. There is no defined limit of believability. It all depends on the sound effect/s in question and is also affected by what other sounds are occurring in the mix. So, it's essentially a matter of experience. B. Laptops, headphones, TVs and even studio monitors all have different output characteristics; certain audio frequencies are amplified, others are attenuated and certain frequencies will be missing all together. In other words, what may sound realistic or believable on a laptop or even on studio monitors will NOT sound the same in a cinema (say at a film festival), where they probably have a $100k+ sound system and the size of the room and speaker placement is orders of magnitude different. There is NO solution to this problem for the no/lo budget filmmaker. Just do the best you can with what you've got and err on the side of caution when pushing those believability boundaries!

G

Digging this thread. So much to learn...

Hi everyone. This is a really interesting and enlightening thread. I'm discovering that 'sound design' is one of the most misunderstood (and poorly-defined) terms used in sound for picture, so it's great to see a discussion of it here.

While reluctant to cause controversy - especially as a 'newbie' here (!) - I'm not sure the definition of 'diegetic' and 'non-diegetic' are quite right. My understanding is that 'diegetic' refers to anything within the 'world' of the narrative, even if happening off-screen. So in the example here, birdsong, background traffic noise etc would still be diegetic, even if not visible to the viewer.

Conversely, 'non-diegetic' refers to sound 'outside' the world of the narrative. So examples would be the film's score, a voiceover/narrator, or certain kinds of sound design elements (like stings or drones in a horror film for example).

My rule of thumb is to ask whether the characters could hear it, or is the audience in on something they're not aware of!

Like I say, that was my understanding, but I could be wrong...

mikemcguill said:

I'm not sure the definition of 'diegetic' and 'non-diegetic' are quite right. My understanding is that 'diegetic' refers to anything within the 'world' of the narrative, even if happening off-screen. So in the example here, birdsong, background traffic noise etc would still be diegetic, even if not visible to the viewer.

Click to expand...

You could be correct, I've only heard diagetic used to describe specifically what you see on screen but that doesn't mean to say those who I've heard use it were correct. I'm thinking though that if diagetic sound were anything which could be in the narrative world, rather than just what you see, pretty much every sound in most films would always be diagetic, with the obvious exception of incidental music. If this is the case, the term doesn't appear to have a great deal of practical use. To be honest the term is still not widely used by experienced filmmakers it's much more an academic term. Most use terms like "hard fx", "soft fx", Foley, "background" and "ambiances" to describe sounds. I never heard the word used in the profession, a few years ago when I became a lecturer for a while is when I first came across the term.

G

AudioPostExpert said:

You could be correct, I've only heard diagetic used to describe specifically what you see on screen but that doesn't mean to say those who I've heard use it were correct. I'm thinking though that if diagetic sound were anything which could be in the narrative world, rather than just what you see, pretty much every sound in most films would always be diagetic, with the obvious exception of incidental music. If this is the case, the term doesn't appear to have a great deal of practical use. To be honest the term is still not widely used by experienced filmmakers it's much more an academic term. Most use terms like "hard fx", "soft fx", Foley, "background" and "ambiances" to describe sounds. I never heard the word used in the profession, a few years ago when I became a lecturer for a while is when I first came across the term.

G

Click to expand...

I never claimed it was useful ... I've heard it used in the term 'diegetic shift' (I think I got that right), which I think is where, for example, a song playing on the radio in the background of one scene seamlessly becomes the background music in the next scene. So it 'shifts' from being in the world to outside it.

But you're right - most people would stare blankly at you I think!

I'm going to resurrect this thread since it is very important information. Here's another quick look at sound design. Pretty basic, but interesting anyway.

http://www.soundesign.info/reviews/the-simple-education-of-sound-design-for-film

APE, Alcove, have you participated to this video ?


"We give our lighting department sometimes hours to set up and then tell our sound team to be ready in 5 minutes".

I'm quoting the video and it sounds so familiar .


https://www.youtube.com/watch?v=BWN3RJGUetk

That's a good little tutorial containing some very good advice/info. However, technically it's off topic because despite it's title, it's got pretty much nothing to do with sound design!! The tutorial deals with fundamental technicalities of production sound recording and a bit about dialogue editing/mixing but doesn't even mention the role of the sound designer, let alone anything about the actual design of sound!? It just goes to demonstrate how poorly sound design is understood by some.

TheArtist said:

It's hard not to analyze the scene from a filmmaker point of view at first but then the tension kicks in and I become just part of the audience. I noticed the room tone and the pen scrapping before being absorbed in the drama.

Click to expand...

I should have picked up on this statement previously for two reasons; firstly, it was the best/nicest compliment given and secondly, it not only goes to the very heart of what sound design is but to the very heart of what narrative filmmaking is, or should be!!

Some films are specifically designed to be a purely visual feast or a purely visual and aural/musical feast, to stimulate/titillate/fascinate the senses of sight and/or hearing. One the best and most influential examples of this type of film is Koyaanisqatsi. However, this goal is usually completely opposite to the goals of most narrative films. For example, when watching Koyaanisqatsi you are struck throughout by the high quality cinematography, which of course is the filmmaker's intention. In a narrative film though, if your audience is thinking about the quality of the cinematography then you have failed!! You need your audience to be absorbed in the story, to feel as if they are there, experiencing the drama unfold, like a sort of invisible participant. If the audience are thinking about how the film was made, the quality of the cinematography, acting, music or whatever, then by definition they are not absorbed/involved in the film itself and the story it's trying to tell! Once down the pub after the film has finished you of course want the audience to reflect and even comment on the greatness/skill of the directing, acting, cinematography, etc., but it's the kiss of death if they are considering these things while actually watching your film!!

As a professional filmmaker for many years I have a natural inclination to analyse a film while I'm watching but sometimes I find this very hard to do. I start analysing but after a period of time I get caught up and adsorbed into the story and realise, sometimes after quite a long time, that I'm watching/experiencing the film rather than analysing it. Sometimes, even if I've seen the film several times previously and I'm watching it specifically to analyse it, I still can't help myself from being manipulated by the filmmakers and absorbed into the story. While it's annoying wasting time/passes and not analysing anything, for me personally, this is the single best sign of a truly great film!!

Ultimately, this is what sound design and filmmaking itself is all about, telling the story in a way which completely absorbs the audience. Audiences don't pay to watch a story or even to be told a story, they pay to be absorbed by and to experience a story, and this is true of ALL forms of story telling, not just film making. So for you to say that you started analysing the example clip and then got absorbed into the drama is the best compliment you could have given me, thanks!

G

mikemcguill said:

I never claimed it was useful ... I've heard it used in the term 'diegetic shift' (I think I got that right), which I think is where, for example, a song playing on the radio in the background of one scene seamlessly becomes the background music in the next scene. So it 'shifts' from being in the world to outside it.

But you're right - most people would stare blankly at you I think!

Click to expand...

I think your description of Diegetic is right, comes from the greek meaning "In Story" I think. So doesn't have to be seen, anything that can be heard by the characters is what I learnt it to mean...

I really like that tutorial/guide posted by Alcove but I thought it started to sound less real as they went on...the bullets and the 'future' sounds jumped a shark for me.

Another little gem. Thanks APE. Planning on reading this in its entirety tomorrow.

All the breathing noises by the victim; were they recorded at a different time? How did you have so much latitude to play with the breathing sounds of the victim? Or did you take the recorded stress diagetic breathing sounds and then place it according to how you guys decided to design it? (not sure about my terminologies).

trueindie said:

How did you have so much latitude to play with the breathing sounds of the victim?

Click to expand...

I'm not sure I understand the question. Once you've read through and watched the example and the first dozen or so posts in the thread you might find that your question has been answered. If not try rephrasing it and I'll try to answer.

G

AudioPostExpert said:

00:08 Just before this cut we introduce some quiet breathing sounds of Marilyn (the victim) to focus the audience's attention on her and help carry us across the cut.

Click to expand...

So this 'quiet breathing' sound, where did you get it? Was it the breathing from the scene that you cut up played with, or did you ADR the breathing at a later time. That was really the question I was trying to ask.

I'm only just trying to understand the process, and what you had to work with, and wondering if you had the opportunity to ADR things and foley things as you decided you needed them


Second question:
I recently saw an indie film where the dialog sound was perfectly clear, and it wasn't ADR dialog, but there was something about the volume level of the dialog that kept putting me off, and as I kept thinking about it, I realized that it felt to me like the sound was coming from my speakers and not from the actor's mouth? I don't know if I'm making any sense, but something about the dialog volume level kept throwing me off. I was wondering if this problem is in my head only, or if you're familiar with it, and if you are, how does one guard against this very subtle off-putting thing?

Thanks.

trueindie said:

So this 'quiet breathing' sound, where did you get it? Was it the breathing from the scene that you cut up played with, or did you ADR the breathing at a later time. That was really the question I was trying to ask. I'm only just trying to understand the process, and what you had to work with, and wondering if you had the opportunity to ADR things and foley things as you decided you needed them

Click to expand...

The breathing sounds came from a number of sources. Some of it was from the production sound but the majority was ADR. In addition to the ADR, a range of breathing sounds were recorded wild (not in sync with the picture) during the ADR sessions.

There is a formal workflow in audio post due to the number of people involved, the amount of work required and the schedule. So it's not really a case of doing ADR and Foley as "you decide you need them" but more like the other way around! The ADR, Foley and Sound FX requirements are planned first, at the beginning of the audio post process. Depending on the desired Sound Design as identified in detailed spotting sessions between the Sound Designer/Supervising Sound Editor and Director, Spotting Lists are created and then the audio post teams (Foley team, SFX team, ADR team) begin creating the materials detailed in these spotting lists.

trueindie said:

I recently saw an indie film where the dialog sound was perfectly clear, and it wasn't ADR dialog, but there was something about the volume level of the dialog that kept putting me off, and as I kept thinking about it, I realized that it felt to me like the sound was coming from my speakers and not from the actor's mouth? ... I was wondering if this problem is in my head only, or if you're familiar with it, and if you are, how does one guard against this very subtle off-putting thing?

Click to expand...

It's a common issue. There are a number of different causes but in most cases it's not actually a volume issue or not solely a volume issue. Most commonly it's an issue of audio perspective, matching the audio perspective of say ADR recorded in a studio with the visual set/location. You seem sure it's not ADR, although I'm not sure how you can be so certain. If it really isn't ADR, the next most likely cause is that the production sound from the lav/s rather than from the boom mic has been used in the final mix and the re-recording mixer hasn't adequately given it the correct audio perspective. There is no way of avoiding this issue, it needs dealing with by a re-recording mixer with the right tools (and experience of using them); an appropriate monitoring system/environment and usually some combination of EQ and reverb (and possibly other tools).

G

AudioPostExpert said:

So it's not really a case of doing ADR and Foley as "you decide you need them" but more like the other way around!

Click to expand...

Got it!

It's a common issue. There are a number of different causes but in most cases it's not actually a volume issue or not solely a volume issue. Most commonly it's an issue of audio perspective, matching the audio perspective of say ADR recorded in a studio with the visual set/location. You seem sure it's not ADR, although I'm not sure how you can be so certain. If it really isn't ADR, the next most likely cause is that the production sound from the lav/s rather than from the boom mic has been used in the final mix and the re-recording mixer hasn't adequately given it the correct audio perspective. There is no way of avoiding this issue, it needs dealing with by a re-recording mixer with the right tools (and experience of using them); an appropriate monitoring system/environment and usually some combination of EQ and reverb (and possibly other tools).
G

Click to expand...

I'm not a 100% sure it's not ADR. I went back and forth in the video to figure out if it could be, and I didn't think it was because the lipsync seemed too good. But yes, the idea of lavs playing havoc with audio perspective sounds like a fantastic explanation.

I recorded boom, and lavs also for backup. One of the issues we were facing during shooting was that our main character spoke in a much softer tone than whoever else she was talking to. So while my audio guy adjusted her lav accordingly he was having some trouble making sure she sounded alright on the boom. We even discussed if we should have the boom pointed just at her (my non-sound person idea of course), but my audio guy advised against it, because it wouldn't sound right. I think he made the right decision of course, now that I'm beginning to see the kinds of things that are possible in audio post.

Thanks for your time APE. Much appreciated.

trueindie said:

Got it!

Click to expand...

Please note that I was answering the question of what we did and of the professional audio post process/workflows. At the lo/no budget level, with no audio post teams to manage and co-ordinate and no experienced Supervising Sound Editor, the pro audio post workflow/approach will probably not be effective!

trueindie said:

I went back and forth in the video to figure out if it could be, and I didn't think it was because the lipsync seemed too good.

Click to expand...

Given decent ADR recordings, decent audio pros and the tools we have available today, it is possible to achieve absolutely perfect lip-sync.

trueindie said:

One of the issues we were facing during shooting was that our main character spoke in a much softer tone than whoever else she was talking to.

Click to expand...

Another occurrence that's entirely common to some degree. Extreme examples exist occasionally even at the top pro level, the "Mumblegate" fiasco with the BBC earlier this year is a prime example. In extreme cases, it's ultimately down to the director to direct the talent to perform in a way which is recordable. Your production sound guy gave you good advice BTW.

G