Being Post-Productive

One of the most difficult aspects of this Jingle assignment was CHOOSING a jingle. After that colossus task, it became the interesting, challenging and enjoyable period of recreating a viable sound to suit a story. 

I have never previously attempted a task such as this, so there has been a lot that I have learnt in the process, about the technical aspects involved, my keenness to partake in post-production and any ways in which I could improve upon my methods in the future.

For my first clip, I went to a film I have long enjoyed and admired, both for the visuals and the produced sound - The Triplets of Belleville - a 2003 animated film directed by Sylvain Chomet.

There is a scene in which the characters perform on stage using various found objects - a vacuum cleaner for example - this is what inspired me to pick this film to work with, as it would be fitting for me to then use foley to reproduce the sounds. Though in the end, I chose a different, but similarly musical, scene, which would give more freedom in composing. The scene is also set outside, so I could experiment with ambient sounds and include a larger range of noises.​

Here is the final product! - 

​I began this assignment by composing the music to the first minute or so of the scene - where the lady and her dog are sitting alone in a somewhat destitute part of town. I muted the original clip as I wanted to produce an original take on the scene, or at least one directly inspired by my personal interaction with the story and visuals. 
​
To begin the scene I wanted an eerie, sombre sound, to tailor to the strangely alien environment of the animation, and represent the ill-fated situation of the character. I thought of a low extending brass sound to form a base for the entire sequence, this was created through the MIDI sound of ‘Soft1 Warm Brass’ (found by auditioning the possible sounds), and played as a C2 note (the C below middle C).

With the boat in this scene, I thought I could also add to the ‘bass-iness’ of the feel by having its horn as an addition to the music. (Please refer to Appendix A for details).
 
With these lower, longer-sustained, almost background sounds, I added a light melody played on MIDI with the 'Flute Lead' preset, to give a more curious depth to the opening visual of the scene, and imply more of the further story.
 
I wanted the sounds to represent something unfortunate yet hopeful - not being too minor but representing a soon to be resolve. The key turned out to be C major (or A minor) as I did not use any sharps or flats. 

I aimed for the music to give a bright intrigue, especially as the old woman began to use the bike wheel to make music. This is where I brought in an accordion sound, I picked this as the feel portrays a light-hearted, playful vibe (and I generally love every piece of music with an accordion in it). Also because the film is set in France, so I thought an accordion could (though cliché) help to set the context of place.
I created this sound through the Accordion preset, under Piano and Keys.

I found I really enjoyed this part of the process; actually composing music to suit the scene. I contemplated having more elements to the melody but decided against it; trying various extra sounds like strings, guitars and violins, even a cello of sorts (initially my intention). It ended up sounding too busy for the scene though, especially because of the extra ambient sounds added, and the light refrains such as her and the dog exhaling.
 
I wanted there to be underlying ‘city’ sounds in the scene, to set the environmental character of where the lady was. For this I recorded a moderately quiet street - to gain some sound of traffic but also to have more distant traffic sounds (to align with her proximity to busy streets in the shots), including the sound of a level-crossing and placid ambience such as birds chirping. I was very happy with how it came out! It sounds natural and does not overpower the scene. I also recorded a siren to add to the hustle-bustle feel of the area which she was in.
 
Along with this was the fire within these scenes that added to the ambience, and her prodding it with a stick. To get the effect of a crackling fire I combined 3 different timbres. I recorded myself slowly crunching paper, and newspaper, and put these together with previous recordings I had of jangling keys. 

​For the woman sighing and the dog sounds I recorded my own voice (with an ATM 4050). For the dog I EQ’d to lower the high-end and boost the ‘gruntiness’ and transposed the pitch in Ableton to -15. I tried to record my dog but he never seemed to make that sound when I needed him to haha

​Once all these sounds were down, it was time to add in the foley for the characters' music-making!
For the bicycle wheel, I thought me tapping a selection of pots and pans would give a sound akin to the metal. 

​I went in the studio with various kitchen appliances and played every part of them I could figure out; producing a wide array of various metallic sounds.
 
I played single hits to use as samples within a Drum Rack, to give flexibility with the sounds, and also played patterns that would fit the footage (for me to insert as audio).
 
There are parts in the scene where she is just moving the wheel around and grabbing it and such, so I recorded myself handling the pots and pans and shuffling them about. 

​These were given EQ seperate to the hits, with a low-pass filter, to lessen the high twangy sounds, a boosts in the lows and mids to accentuate their slow, subtle timbres.

---

Something I did not consider in all my excitement was how challenging it would be to match these parts up with the footage. Or not even that it was difficult so much as it was time-consuming! I had to quite meticulously arrange and rearrange the parts until it was convincing that they were coming from the actions on the screen. 
​
​Although, in the film itself, much of the music is not in sync with the character making it, so I found a balance by syncing the music and visuals up when it could fit but also loosely dancing around the beats when necessary, to still create some sense of accordance whilst also having freedom with what music I created.

When the three ‘triplets’ walk in scene, I wanted a mysterious, luring sound, and thought it would be cool to use the ‘Cosmos Soundscape’ for this part. I also used another ‘Ambient & Evolving’ sound which comes in to integrate with the first - the Simpler ‘Chill Outzone’. At the same time as the first character comes in shot I played the ‘Solo Female1 Fast Attack’ Operator - I absolutely love this sound and as soon as I heard it I placed it to come in with the three curiously ephemeral characters, coupled with the footsteps I think it has the perfect spectral feel to represent their presence.

​With the footsteps I used samples I recorded previously for the sample library, with decent reverb (a large hall with decay time of 4.2 seconds), and made a separate footstep track for each character; panning them to suit where each were in the shot.
 
From this point onwards, it was time for intense sound replacement and beat-building!
 
For the first part of their singing, I focused on sound replacement. I created a body percussion drum rack - with samples I recorded of me hitting parts of my body, clicking and clapping, along with the footsteps, jumping, stomping, foot slides (recorded in a tub full of small stones) and steps crunching down on stones to emulate gravel.

---

The singing, was an enjoyable undertaking of my friend, sister and I. (Please refer to Appendix A for more details on the recording and editing).
 
I continued using the two drum racks here (one of the pots and pans and the other of the body percussion) to create the sounds that the characters were visually making - to a certain extent. I picked out the important aspects of the characters’ visual movements to match, but then focused on having some creativity in building my own rhythm, whilst also keeping credible that what is on the screen is expressing this music.
 
Lastly in this project were the laughs. This was the most enjoyable part of the whole process - recording them and putting them together felt ridiculous. I mostly left them natural, with slight EQ to take out any ear-piercing frequencies, but added a heavy reverb, of a Large Church with a 6.3 decay time (and increased the reverb as the camera view rose up past the bridge). I also wanted the laughs to hang over the footage, and that’s probably my favourite part of the whole clip.
 
Overall I am quite happy with it! I enjoy watching it, knowing how much work I put into it, and seeing that it came together as a full product!

​After all of this pain-staking sound-replacement, the next clip I chose had (purposefully) quite a different process of working to it. I picked the final sequence of the 1920 film ‘One Week’, starring one of my favourite actors, Buster Keaton (directed by Keaton and Edward F. Cline).

​The music in itself I am not hugely proud of, it was the second video I focused on and became evident as the ‘second’ part of this assignment… But I am quite pleased with the result in a score point of view – as in, with its relation to the clip. I believe it expresses the sentiments of the clip satisfactorily, so for that I am pleased.
​
Here is my version of the scene - 

 
I enjoy silent films greatly, and am in awe of how the actors and directors convey a story without dialogue. I thought it would be an interesting task then, for me, to add to this visual story-telling with only composed music – no sound effects, no foley and no dialogue. I have never tried this before so I wasn’t sure what to expect, and it did prove itself quite challenging!

Overall for this piece, I decided I wouldn’t nitpick with timing and tempo, but rather, play along with film, and focus on feel and style of the 'real' playing (as real as a laptop keyboard could be considered). If any parts were quite obviously off though I fixed them, but I wanted the piece to be quite fluid, free-flowing and natural - not over-produced or electronic sounding (even though the 'instruments' are algorithms). I guess that was the challenge!
 
The most difficult part for me was the layering of the instruments - using certain instrumental combinations for different sections, that fit in with particulars of the story and at the same time sounded pleasant in junction.
 
I started composing with a piano; intending for that to give the main melody of the piece. I spent time just improvising short melodic sequences for each part of the film, and chose the ones I liked most; having one or two main melodic sequences to carry the entire piece (I didn’t want randomness the whole way through but a certain unification of sound).
 
The piano alone was sounding slightly uninteresting, so I consulted Ableton’s Instrument Rack for some more music flavours. I decided to stick to orchestral instruments, the keep a certain honesty to the silent film era, and keep a consistency between the visual and audio. (Please refer to Appendix B for instruments used).
   
I found with the amount of instruments I initially aimed for, there was still a strange hollow-ness – although I wanted it to be simple, and quite minimalistic, there was still depth missing. I added some extra parts, to cover a greater range of frequencies – providing low parts through a bassoon and tuba, through to a piccolo at the high end.
 
It was starting to come together more with each added aspect, and editing of the distinct parts to have a nice interplay between instruments. I found myself deleting MIDI notes a lot, taking some away from certain instruments to allow others to come through. It was all about contrast and connectivity for me.
 
At the very end, there was still something missing. I tried adding percussion and think it completed it! Even though I used it quite sporadically, it added another element to mix, and helped express some of the comedic value in certain parts.

​
Doing this activity further cemented my willingness to compose music for film! This was highly gratifying for me! I love the connection between the visual and the audio and this is one of the most interesting and satisfying ways of carrying this out.
 
Both clip feature on ViVE’s company wesite, at http://viveproductions.weebly.com/

​

​APPENDIX A
 
The boat sound:
 
I used a field recording (from an iPhone) of a train horn - this I EQ’d to increase the low frequencies, and also pitch-shifted in Ableton (to -30). I warped each horn to different segment bpm’s (higher to make it longer, shorter, quicker), to create divergence in the sounds of the horns, making them more realistic.
To add depth to the horn, and make it an odder sound, I layered with it a synthesiser (called ‘Cosmos Soundscape’), that had a portion of Modulation and gave a vibrato to the sound. I played around with the tension and available parameters to get a sound that represented this quirky imaginary land’s boat sound. I played it at C1 (to keep consistent with the brass note).
 
The fire sound:
 
With the paper track, I EQ’d them in Pro Tools to heighten their ‘crack’ and ‘crunch’ (at around 2-3kHz), and the high end. I also found around 150Hz gave more depth to the crunch.

​I sent all the parts to the same Reverb track, with the below parameters:

 
Pots and Pans:
 
With all these clips, I EQ’d them in ProTools, applied Compression and sent them to a Reverb track, along with a SansAmp plug-in, to give a trashier, scruffier feel to the tracks (making it believable they’re the sounds of a dingy bicycle wheel found under a bridge).

​Shuffling sounds:
 
These were given EQ separate to the hits, with a low-pass filter, to lessen the high twangy sounds, a boosts in the lows and mids to accentuate their slow, subtle timbres.

​The Dog sound editing:

​The Singing:
 
For the sections where all three women sing together, I used an ATM4050 on Omni-directional, with the three of us standing around the mic in even spacing. We tested out the sound, our distances, and found a spot where I was happy with the balance! For the ‘verse’-type part, I wrote what I believed were the lyrics and we listened for the original harmonies to each separately sing a part each (low, medium and high), using the same microphone but on Cardioid pattern. I also added an extra high part on the last words of each line (I panned each part these to give each layer of the harmony their own space).
 
The processing of the vocals included corrective EQ, which was different for every track (mainly to take out any annoyingly high and/or sibilant frequencies and nasal-ness) and Compression. Plus for the individual parts, I thought they were not raw enough - they sounded too clean and crisp - which is usually a good thing, but I didn’t think they suited the character of the animation. So I sent them to a track with SansAmp to add some crunch and buzz; making them seem a little more vintage.
 
After this section there is a part in the music where it changes and becomes thicker and heavier and has more of a groove to it. The vocals here I wanted very grainy and lo-fi, so, after testing various effects, I applied the Eleven Lite plug-in. It gave the voices a good old-school, somewhat Blues feel, which is what I wanted! I found this wasn’t extreme enough for the solo part however, so I used that as well as a SansAmp.

​​The solo was funny to try to replicate - the character is just projecting gibberish, so I started doing exactly that, but then I kept incidentally saying certain lyrics, so I wrote them down and voilà I had a song!
 
It was sounding too clear though and I wanted to keep some abstraction, so I layered a take with gibberish under this one, and pitch-shifted it down a semitone, and panned it out slightly, to give it some disparity.
 
With the background vocals here, the two singers recorded this together, and I duplicated this clip to pan them hard left and hard right, to underlie the ‘solo’.

​

​APPENDIX B

‘One Week’ session

Instrument tracks used: