Sound Design Report

The task I was given was to re-design the sound for a scene from a film. The scene had to be around three minutes long. The point of the task was to use the range of recording and editing skills developed in the sound workshops to create an interesting and original piece of sound design.

I chose the scene from Only God Forgives because I think the sound does not suggest to the audience a quite important piece of information; the main character is fantasising not experiencing reality. I understand that this is the director's intention but I still feel that the scene would benefit from the  sound point out this piece of information. My basic plan was to create two clear sections to the sound design; a reality section and a fantasy section.

I then divided the scene into sections and placed markers in Pro Tools at the in and out points of the sections.

After researching Cliff Martinez's film score techniques, I wrote a piece of music that would fit the mood of the film. The piece consisted of simple chords played on a MIDI keyboard with the Accordion In Space software instrument in the Xpand!2 Pro Tools plug-in, I also adjusted attack, decay, release, cutoff, env. depth and fine tune in Xpand!2 to the create the desired effect. I also used the Vintage Filter plug-in to add modulation to the MIDI track.

To distinguish between the different sections in the scene, I made the levels of the music during the fantasy sections higher than they are in the reality section. This was to add to realism of the reality section and to add to the dream-like quality of the fantasy section.

The next step was to record sounds that would be needed to aid the actions on screen, for example footsteps or clothing movements. I started this process by watching through the clip a number of times and writing down the sounds that needed to be created. I knew I wouldn't be able to make a few of the sounds just by doing what is done on screen. For example, the slap towards the end of the scene is very aggressive and I wouldn't be able to slap someone hard, so I came up with solutions to these problems such as, gently slapping a number of things and planning on layering them in post-production. Another problem I had was creating a sound for muscles flexing as there isn't really any sound made in real life, so I tested stretching and twisting a few different materials, the best sounding one was leather so I recorded leather being twisted for the sound of muscles being flexed.

Recording an atmosphere track proved difficult as I was wanting something that would fit the visuals of the scene. The scene is set in a seedy bar in Bangkok which is obviously different to Sheffield. Another thing that I wanted in the atmosphere track was people talking in a bar environment without the music, this also proved difficult. After recording a number of atmosphere tracks, I settled with the one that had the most human conversation and least music in the background.

During the recording process I experimented with certain sounds that could be used for ambience. I recorded the sound of my toilet's flush running continually, I recorded the microphone being gently dragged along the side of my radiator and I recorded my watch ticking. In the end I used using all of these sounds in the final piece, the most prominent being the watch ticking towards the end of the scene creating tension.

The editing process was fairly 'straight forward' the main task was deciding which sounds to put where and for how long. Once I had placed the sounds on tracks on Pro Tools, I altered their levels so that they sounded right in the mix. There were some sounds that required reverb and EQing, I did this with the plug-ins EQ3 7-Band and D-Verb.

 The final thing to do once all of the levels were right was to create a master track and change the master level of the piece.

I am happy with the final piece, I feel that the outcomes have, on the whole, satisfied my aims. The two sections of the scene are more prominent now, I feel. The main success of the task for me is the fact that I have created a believable soundscape that adds to the mood of the visuals.

The diegetic sounds that are in the scene work well for the most part. I have used very subtle diegetic sounds as well as very unsubtle diegetic sounds. The subtle sounds in the scene are more successful as when recording they were more controllable, for example, when recording the beads jangling, it was easy to follow what was on screen but when I was recording the sound of a slap, although it was easy to sync with the visual, it wasn't easy to recreate the exact sound.

The music in the film, I feel fits well and draws the attention of the audience towards the action on screen. The sections in which the music track is higher in the mix are more engaging and the jump up to that louder level is obvious. Whether or not the reason for the change in levels is obvious, I don't know but I feel that there's enough of a change to be asking the audience questions about why it's there.

The two sections of the film are well distinguished in the scene. In the reality sections of the scene the EQ of the atmospheric track gives it a more present and real feel whereas the music is the secondary sound in the section. In the fantasy sections of the scene the music is the primary sound and the atmosphere track is more distant giving a more dream-like effect. The only exception to this dream-like effect was the fact that during the fantasy sections, the 'real' sounds such at the beads were quite high in the mix and literal, I feel that this maybe stumped the fantasy feel, however I think that it worked in the scene as a whole, as without them the scene may have felt a little boring and unreal throughout.

If I were to do the task again I would do a number of things to improve the piece. Although I am happy with the piece as a whole I think that certain parts are not believable or 'cheap'. The main changes I would make would be the overall production values of the piece and the attention to authenticity.

The atmosphere track in my piece is, to an extent, not authentic. The low, underlying bass-line in the recording does not fully fit the mood of the piece, it sounds too western and happy for the scene. Also, the majority of voices in the recording are British accents, again not authentic. The sound quality was compromised a bit as I recorded the stereo track directly into the Zoom H4N's microphone as it was subtle and I didn't have permission to record sound in the bar. If I were to do this task again, I would scout a better location for the atmosphere track, i.e. a Thai influenced bar. Another thing I would do is ask for permission before recording, allowing me to use a larger and better stereo microphone, giving a better sound quality.

The vocalisations in the film (grunts from bodybuilders, female exhaling and men in pain) were all recordings of me. I feel that this may have been obvious throughout the scene, even though I EQ'd them added reverb to them. As well as all of the vocalisations sounding similar, they weren't delivered brilliantly which is maybe also obvious. If I were to do the task again I would find a number of capable people to perform the vocalisations for me, this would give a range of vocalisations performed in a more convincing manner.

Although the music that I wrote for the piece worked well, I feel that there was room for improvement with it, there wasn't a lot of depth to the music as it was just a single software instrument. If I was to do this task again I would create more texture in the music by experimenting with more software instruments alongside each other. I would also experiment with real instruments and ambient recordings to add to the texture of the music.

Overall I am pleased with the end result. I feel that the two sections of the film are clearly defined. Also, I feel that the sounds that I have created, wether they are solo sounds, layered sounds or MIDI sounds, fit well with the visuals and sound realistic with the visuals. Although there are some aspects of sound that could be improved I feel that I have designed the sound in a way that fulfils my aims with a good degree of understanding. I feel that the sound is believable and done well enough to define the two clear sections of the scene.