Search
  • Ryan van Eerde

Making Pirate Sounds


Today I'm going to be discussing how I produced some of the sound design elements in my Pirates of the Caribbean sound replacement project based of the research I have talked about previously.


All of the sounds that I talk about today were recorded with a Rode K2 microphone, why because its a great sounding mic with low noise so if any recorded clips need gain boosting there will be minimal noise boosted as well making sure everything is crisp and clear. Read this review for more information http://recordinghacks.com/reviews/rode-k2-sings-like-an-angel/ .


Glass Break


In my research I found a technique of through a key into a glass to produce a glass break sound. I used 3 different sized glasses and a couple of different keys in order to produce a number of different sounds that could be blended together to produce the texture and tone that I wanted. I did multiple takes both with and without cling wrap over the top of them (see previous blog for more details). My project partner Dustin found in the small Foley studio we were using a cardboard box filled with broken plates which coincidentally was another technique I researched to produce glass break sounds. So, seeing as we had this at our disposal we recorded a few sounds of me hitting the box full of broken plates as well because you can never have too many sounds.


After they were all recorded I edited the different takes into singular hits that could be copied and moved around in the sound design process, see image below.

From top to bottom, glass with cling wrap, glass without cling wrap, box of broken plates.

I also cut out a lot of sounds that I deemed to be to usable and adjusted the clip gain on most of the sounds to get them all to be roughly at the same level.


Next a went through a process of choosing what were the better sounds out of them all to produce the most realistic glass break sounds. I tried to go for the ones that sounded like they would need the least amount of processing and could almost be considered a convincing sounds by themselves, see image below.


The best sounding clips and again from top to bottom, glass with cling wrap, glass without cling wrap, box of broken plates.

From there it was just a matter of layering all of the clips together and make a couple of different glass break sounds, see image below for the layering of the different clips. I had all of the audio tracks running into a bus that then gets sent out to a print track, I did this so I could process all of the clips in the same way and record them down to one stereo audio track.

The production of 4 different glass breaks and the layering of clips to produce them.

The first glass break is meant to be a quick sound with more of an initial impact and not really having anything happening at the end. I basically lined up the clips so that all of the main hits are played at the same time to really emphasise the impact of the glass break. I also did some pitch shifting to the last 3 clips to give me the right tone ( +9, +7, -2 semitones from top to bottom).


The second glass break was much the same as the first but I staggered a couple of the clips to give me a little bit longer of a break/smash sound that is also denser which could be used to represent a bigger glass. This time I only pitch shifted the bottom clip +2 semitones because that's all what was needed.


The third glass break is meant to be much longer with the aim of making it sound like multiple piece of glass falling and crashing into each other. To do this I staggered the different clips more severely than the second one and also time stretched the last clip by over 100% to give me a really slow and long jangly key sound to help tie all of the different clips together. I did pitch shifting on most of the clips between +9 and -4 semitones (I'm not going to list all of them for this one).


The fourth and last glass break is a slightly longer sound of the first glass break with a couple more clips to the end to fill out the sonic content a little more. I time stretched the first 3 clips (78%, 62%, 43% from top to bottom) to help make the break sound like it was a little longer but I also wanted a couple of normal clips in there so it wouldn't sound too processed.


Going over all of these sounds is some EQ and Compression, see image below.

Eq and Compression settings.

I basically just used the EQ to reduce some of the frequencies that made the glass breaks sound a little too metallic and boosted the high end to make them all sound more glassy and shimmery. I used the Compressor to even out the dynamics slightly to help glue all of the different clips together and make them sound more like they all came form the same sound source.


Once I was happy with how everything sounded I recorded the clips down to stereo audio files to be imported into the main production session.




Gunshot


For the gunshot I recorded popping 4 plastic bags to give me a good bang sound that I could process later to make them sound more like musket fire. I also used a SM57 dynamic microphone along with the Rode K2 capture the sounds to see what tonal differences if any it would pick up, see image below.

The 4 bag pops with the top track from the K2 and the bottom the 57

Looking at the wave forms from the 2 different mics its visually very interesting to see how they respond to the source sound and just how much more information the K2 pics up. This is a good visual representation of what different microphones can do and why choosing ones to do specific jobs is paramount for capturing sounds correctly, but that's a blog for another day.


To make the sound I layered 5 of the bag pops together then did some relatively extreme time stretching to the first 4, see image below. Unlike with the glass breath sounds where I used the Polyphonic elastic properties mode that keeps the same pitch when time stretching I used Varispeed here which means when I make clips longer they get lower in pitch. Also when you time stretch with the Polyphonic setting the clips can only go so far before that start sounding glitchy and not natural where the Varispeed can go a lot further. Going down the clips I stretched them out by 214%, 118%, 165%, and 239%, I basically just kept stretching them until they reached a pitch and length that I liked. The very last clip has been completely unprocessed to provide more of a quick bang sound.

Layering of the 5 different clips.

Like the glass break I had all of the clips going into a bus then to a print track, but I also sent all of the clips to 2 more Aux tracks, one with a oneknob driver plugin and another with a reverb and Eq, see image below.

Plugins used on the different Aux tracks

The first Eq was used on the main Aux track to help shape the overall tone and make it sound a little more. The second Aux with the oneknob driver was blended in to add a bit of saturation and distortion to the main sound to make it fuller and more distorted obviously. Lastly the reverb to make it sound bigger and wider whilst also helping to make it sound like its inside a room, then a bit of filtering Eq to shape the room tone to my liking. Then I blended all of the Aux tracks together until I got a sound I was happy with the recorded in down to a stereo audio track.



Ricochet


For the ricochet I recorded a hitting a metal pan with a couple of different pieces of metal to produce a hit sound that also ringed out a little rather that just a short clang/hit sound. I also recorded this with the SM57 and Rode K2 combo, see image below.

2 hit tracks, top K2, bottom 57.

After listening through both tracks I picked a few of the best hit sounds which were all from the K2 microphone as it picked up a lot more of the metal ring sound compared to the 57, it also has a lot more clarity (who would have thought), see image below.

Best hit sounds for the ricochet.

I created a couple of different ricochet sounds by layering 3 of these metal hits together, see image below. For the first ricochet I pitched shifted the first 2 clips down by 15, and 8 semitones to make them sound heavier and less like a small frying pan, the last one I kept the same to have a little of the high pitch ting sound in there still. For the second one I used different hits and pitch shifted them all down by -5, -10, and -4 semitones, this was to make a sound that felt like it can a different weight to the first one. What I mean by this is, when you pitch shift something down the lower pitch makes it sound like a denser and heavier object having a slightly different contrast in sounds will help to represent a bullet hit different surfaces.

Different clips used to produce the 2 ricochet sounds, with a stereo and mono sounding recording which I'll get to later.

I used the same processing for both sounds Eq on the first track then the Doppler plugin on the other 2, see image below. With the Eq I reduced a lot of the low end because it had the low frequency undertone when pitch shifted that I didn't like that much, I also reduced 500Hz and 2KHz to help the high end ring come through better.


For the Doppler plugins I used the same initial preset which was a Superfast Flyby, but I slightly adjusted the pitch change and panning of each of them to produce a more interesting and dynamic sound. The preset setting by themselves work well but changing a couple of things between them makes the Doppler effect seem more realistic.


After that I put a compressor of the main bus with a slow attack and release to squash the tail of the sound whilst letting all of the initial attack and dynamics come through, squashing the end makes it appear more like the bullet is going passed you and produces a tighter sound.

Plugins used for the ricochet sound.

Once I was happy with how everything sounded I recorded both of them down to stereo tracks, but I also recorded another set where I changed the panning of the main bus to mono to get a mono sound. I did this just to give me more creative options for implementing the sounds into my main project session if needed, this can be seen in a couple of images above.



All up I think I have produced some relatively good sounds that I believe will convincing when they are in the main project session. They weren't as tricky to produce as what I first thought but I will still need to see how well they fit into the session and then potentially go back and fix some of the sounds if needed. I also have another session where I did some sound designs that I will cover at a later date so stay tuned for that one.













4 views0 comments