Planning & Production

This page details my main process towards achieving my final product. It documents the main timeline of what I did for my practical work throughout the project. Certain specific details may be reserved to the problem solving page where applicable.

Initial Stages and Setting Up my Practical Workflow

In terms of practical work, my first task was to choose a suitable video clip to redub. Referring back to my Problem Solving page, this wasn't easy. More details of this are on the Problem Solving page but to reiterate:

The Clip MUST:

Contain action that requires the recording of practical Foley.
Contain the potential to host a variety of different sound effects and nothing too repetitive to make.
Contain a scene with walking, preferably on a natural surface.

The Clip SHOULD NOT:

Contain any of the original audio.
Contain dialogue.
Be too long (5 minute cap).
Contain anything that I wouldn't be able to record with my timeframe.

As explained on my Problem Solving page, I ended up deciding I want to redub a series of movie clips that had been condensed into a short form compilation clip. The scenes I ended up choosing were from Toy Story 2, Wallace and Gromit: The Curse of the Were Rabbit, and Lord of The Rings: Fellowship of the Ring. Below is a list of the original clips and my reasons for choosing them.

Toy Story 2
"Fixing Woody"

The first scene I've chosen to dub is this scene from Toy Story 2. I chose it due to the very tactile nature of the scene, involving a theme of delicacy and a range of different and specific small objects too. The clip offers a large variety of opportunities to record different sound effects. It isn't heavily dialogue focused either and I can easily edit any remaining dialogue out.

To edit and cut down the video clips into one video I used Microsoft Clipchamp, a very rudimentary form of video editing software that came with my copy of Windows at home. In the professional industry, I would've been more likely to use software such as Adobe Premiere Pro but programs such as these are unavailable to me due to their paywalls. For the purposes of this project (an audio focused project), Clipchamp was suitable enough and easy enough to learn and control to combine my clips together without losing quality. I could even add preset titles to separate each movie clip. The final edited video is linked below with the original audio. I want to have a version with the original audio on hand so I can compare it to my version with the audio I've produced.

Note: I later noticed the typo (a missing apostrophe) in one of my title cards and this slight imperfection is really annoying me. I might fix it later, but probably not as this project's focus is sound, not video. But I am aware.

The next stage in my process basically involved noting down every different sound that needed to be recorded. I explored this process, known as "spotting", in my research. It involved making a spotting sheet in Google Sheets and using it to note down very sound effect that occurs and what specific time, using the film industry standard time code format, SMPTE. It also lists other information such as detailed effects descriptions and sources for each sound. I used a video editing software called Wondershare Filmora (why I had to use this is listed in the Promblem Solving section), to go through my entire video clip frame by frame labelling different times where certain sound effects occurred using the SMPTE time code format.

This was an extremely lengthy process with each 30 seconds of the video taking over an hour to label. I also modified my spotting sheet to include extra information specific to my project. This includes the fact I split the "sound effect source" category into a "proposed" sound effect source, which is a Foley recording from my creative intuition that I think will work for the on-screen action, and a "researched" sound effect source, for Foley I wasn't sure how to make and needed to find a "recipe" for. Despite this, some of my extra categories did go unused for the most part.

After seeing the sheer amount of recordings I needed for the I was extremely concerned about whether I could record all the sound effects in time for the end of the project. I only had a set number of hours in the studio after all. As part this concern, I decided to add a "Recycle?" section where I could label sound effects that could be unnoticeably reused or repeated in the project. One of example of this is reusing the chopping sound effect for when Gromit chops up the carrots. As a last resort I could use externally sourced sound effects (which isn't uncommon when producing sound effects for movies - how many times have you heard the "Wilhelm scream"?), but I wanted to avoid that if I could.

Either way, I now had a solid plan of action for recording my work.

Click to see the full spotting sheet

(some "researched" entries are in the incorrect column)

Recording Stages

I want to start this section my noting that I won't be going into full detail of every single sound effect I record and edit. As seen from the spotting sheet, there are many, many different individual sound effects used in my project. Describing in a satisfactory amount of detail each and every single one will make this project extremely time consuming (and time is limited) and therefore I will show highlights of different sound effects and the recording techniques I used for them. Additionally, some of the techniques used on one object would be similar to the techniques used on another object so I could end up repeating myself, which isn't an efficient use of time or word-space. This isn't saying that this section will lack content. I will post a gallery showing most of the recorded effects near the end.

Setting up Logic Pro

My first task before recording was to set up Logic Pro so I can start efficiently working with it. I started by importing my clips into the DAW. This was a simple drag-and-drop process and it had its own tracks on the timeline; one for video and one for audio. I immediately deleted the audio track, as intended.

Logic's default timing format will count the number of bars and beats per bar along with a regular format clock but, as deduced from my research, this was not the industry standard for film and media production which uses the SMPTE timecode format. Also my spotting sheet used SMPTE. Luckily, Logic Pro supports this and I was able to change the timing display format in the project settings. This was very easy to do, I just had to make sure it matched the frame rate of my video - 30 fps. I could now easily match the placement of sound effects in Logic Pro to what I noted on my spotting sheet. I was now ready to record.

Recording Foley - The First Sessions

Honestly, in terms of foley materials, I was ill prepared for my first recording session. Despite this, I was lucky because the first few seconds of my first scene from Toy Story 2 didn't require many complicated objects. I referred to my spotting sheet which told me the first sound effects were a doorbell, a door opening, then many sound effects involving an elaborate wooden box. I hadn't researched what I should use for these effects yet but in the studio, I managed to find an empty microphone case which looked almost perfect for some of the sound effects I needed.

From my research, I had deduced I would need to use a condenser mic as it had suitable properties for recording foley such as the option of having a hypercardioid pattern which is good for focusing on the target being recorded, rejecting input from rear angles. Their great dynamic range and detail focus would also be a positive when recording a variety of effects. So with this knowledge, I set up an AKG C-1000 with a close mic setup to help directly capture my sound source and minimise any ambient noise from the room.

As this was a condenser mic, after I plugged it in I needed to apply phantom power (48V). For each sound effect I wanted, I would create a new audio track in Logic Pro (naming them too for organisation) and begin trying to create the desired effect. I would first adjust the gain depending on the signal strength of the output; too high output meant I needed less gain, and too low of an output meant I needed more gain. I had learned this from my research on gain staging (see Research - Microphones section). Using this knowledge, I knew I would need to aim for about -18db on the peak meter for each recording as this would help achieve a consistent volume throughout all my project's recordings which helps me when balancing the mix during editing for the final product as every recording will sit around the same volume in the DAW with reasonable headroom for mastering. This, overall will help me achieve a professional sounding final product, which is my aim for the project.

Gain levels would be different depending on what I record because different objects will be louder/quieter than others. For example, in this first recording, the aforementioned microphone case, I recorded both the hinges of the box and the clips of the box for the clips. The clips were a lot louder than the hinges and therefore needed a lot less gain to stay around the -18db mark.

With prep work for the recording done, I would watch the clip where the audio occurs one more time to analyse the movements with what I'm recording (practice generally helped create a better sound effect). I'd then record myself performing the sound effect with the object whilst watching the clip (as long as the video feed in Logic hadn't frozen - see problem solving section for more information). I'd usually try performing the effect in question several times during a recording as it gave me chance to experiment with how the object could sound and more options when editing it later.

When I was satisfied with a take, I'd stop recording and listen back to the recording I just made. I could then trim down to the part I wanted using the scissors tool. Using the SMPTE time codes, my spotting sheet, and Logic's video feed (if it was working), I would then align the sound effect to where it occurred. Getting the timing right was very important as desynchronised audio would make the audience lose their immersion meaning I wouldn't have produced something of a professional quality.

From here, I could also adjust the EQ, and add effects in the mixer rack. From my research I learned that creative use of these tools can help your sound more accurately depict what's happening on screen. For example, the metal box I was using was actually supposed to be mimicking the sound of a wooden box. So, I needed to make the audience immersed and believe they're listening to a wooden box, not a metal one. To do this I slightly adjusted the EQ of the box's recordings by raising the gain of some of the lower frequency bands and lowering the gain of some of the higher frequency bands. I did this out of my own creative initiative as I was aware that metal tends to create more high end frequencies than wood. I thought about the differences between a marimba and a vibraphone. Marimba is wooden, vibraphone is metal, marimba accentuates lower tones, vibraphone accentuates higher tones. With this logic, adjusting the EQ in the way I did worked to a very convincing manner.

For each of the sound effects I recorded (this box and every sound effect hereafter), I added a slight fade in and out using the fade tool. This fade varied depending on the sound effect but usually I would go for around 25ms of fade. I did this as my research had informed me that this would prevent popping when a sound enters/leaves the mix. Getting rid of these pops would help me develop a more professional sounding piece of work, one of my penultimate goals for the project.

Continuing with the first clip, I decided to next tackle the opening of the door in the clip. Again, all's I had available to me at the time was what was around me in the studio. For this, I thought about my research on layering and saw that the studio I was in had a door and a metal cabinet. Each of these on their own probably wouldn't have produced a sound that would convince the audience that the door they were hearing was the one on screen so I decided to record their sounds individually and layer them in the edit. Again, the process was similar here to the box. I had to set up my area for the recording (in this case, setting the microphones up in a close position to the door/metal latch), practice the actions, record into logic, and cut and fade using the fade and scissors tools respectfully. This is basically how most of the Foley was initially made (not including the ambience sounds or individual post effects). I'm not going to keep repeating this for every sound effect I list on this page so assume that's the process taken for each effect.

After the two door mechanisms were recorded, sliced, faded, and positioned correctly in the timeline, I then automated the panning. I learned about automation in my research. The automatic panning here was used here (and actually previously with the box in slight increments) to portray the movement of the door as it gets swung open left to right. To match that, I automated these sounds to pan left to right along with the movement of the door. Using panning in Foley is another great technique to immerse the audience in the media, convincing them the sounds they hear are coming from what's on screen. Creating an immersive piece of work again aligns with my

project's aims. Additionally, I slightly pitched down the metal hinge seen above on the left and lightly boosted a small range of lower frequencies in its EQ. This was to match the non-metallic bigger door seen on screen.

A combination of layering, my editing techniques, automation, and effects usage, produced a convincing moment in my Foley work of a high quality that I was proud of.

Further Sessions with Foley: A Brief Summary and Project Structure

As mentioned earlier, going through each sound effect in as much detail as I previously have is going to be difficult. I'm writing this paragraph after my final studio session and I've got approximately 75 different sound effects. Writing about every single one in full detail will be repetitive and time consuming. Most of the sources can be seen on the spotting sheet too. All of these were made using the first primary few steps:

Set up microphone positioning for recorded object
Practice Foley action and gain stage
Record Foley action(s)
Use scissors tool then fade
Repeat for layering if needed
Position correctly in timeline referring to spotting sheet if needed
Use EQ, pitch shift, and other effects to taste

I would additionally be adjusting the volumes of different sound effects for a balanced mix as more sound effects were recorded and made sure everything sounded the correct volume for the final product and not too loud. All these steps were taken to ensure a high quality final product with the only main difference being what was being recorded and the slight adaptations for that object. From my evaluation, and peer feedback I think this was achieved and the methodology I used to get there was very good. I made some mistakes along the way no doubt, but I've been learning from them and improving myself as a result. From here, I want to continue this section of my project by describing my process recording ambient sounds (as it differs the most from the rest), and then I'm going to show highlights of my favourite moments of my Foley sessions. I think this is the best way to display my project's process and I think showing the bits I had most fun is most enjoyable.

Ambience: A Trip to the Woods

After my first one or two recording sessions, it was Easter Break which meant I lost two weeks of studio time. But, I borrowed a Roland Edirol R-09HR ambient recorder from college to use to record ambience to be later used in my Foley project in the Lord of the Rings clip. This process was a lot of fun. I went to Cannock Chase, a forest owned and run by the forestry commission, and basically walked into the woods to see (or listen?) to what I could record. I plugged in some in ear monitors so I could listen to what was being recorded. This was quite an experience in itself as it basically amplified my hearing capability to a vast degree, letting me hear birdsong in much greater detail. It was very relaxing.

The R-09HR had a stereo mic setup, which produced a very wide, directional output with up to 24 bit 48kHz resolution. Very good in my opinion. But, I want to note, in a professional environment, a shotgun mic setup may be preferred for recording individual birdsong due to it's focused design (see research section). Unfortunately, these weren't available at college.

In terms of the actual recording, I first had to adjust the gain. This was tricky as I struggled reading the R-09HR's dimly lit screen in the sunlight. The earphones I were using also had quite a low resistance too meaning I had to be careful that my final recording wasn't too quiet. This amount of limited control almost felt like a shot in the dark in terms of whether my final recording was going to be good quality.

On top of this, I had to walk quite a long way, getting away from nearby roads (as I didn't want cars in my final recording). The positioning of the trees affected this too, as the trees would reduce the vibrations of the road getting to the microphones if I walked in a particular direction.

After walking for a while, away from roads, dog walkers, and mountain bikers, I found a spot with a decent amount of birdsong, rustling of the tree leaves, and wind, which would provide an immersive ambient feel. I pressed record, making sure to stand perfectly still and watch how I even breathe. The microphones were quite sensitive and would pick up my clothes rustling, me shuffling or my breath. I stood dead still for about 3-4 minutes. I did this about 3 times in 3 different locations. My third location was probably the best as it had the most bird activity and not too much wind peaking the microphones. This is what's heard in the final clip of my project.

Additionally, I had the thought of recording Foley footsteps in the nearby foliage. This was highly unconventional as usually (and as seen in my research), a Foley pit would be used for this. I was unsure if I was going to get time to build a Foley pit at that time so I did it as a backup. Additionally, these footsteps would be layered with the same ambience so only the footsteps themselves would be accentuated in the final project. I recorded my Dad walking over some leaves and this too was used in the final clip (because I unfortunately never got around to building the foley pit) when Frodo walks in the forest. I carefully had to use the scissors and fade tools to edit them in at the right time during the final edit.

These recordings were very good quality in my opinion in the final clip and really help immerse the audience, making them believe they are in the forest themselves. The volume turned out to be fine too. Providing immersion this way has helped me achieve my project aims.

Wallace & Gromit Highlights: Lots of Carrots

The highlight I want to take from the Wallace and Gromit clip is the Foley I did involving carrots. I needed the sound effects for chopping up carrots, dropping them into a chute, and the bunnies munching on them.

For the carrot chopping I brought in a chopping board, a carrot, and a tutor kindly provided a sharp enough knife. My tutor had to bring and use the knife so I didn't get flagged as a security risk the moment I stepped in college that morning. Me explaining in security's office what Foley is would be a waste of studio time too. Jokes aside, I also was using a different microphone at this point, the Neumann TLM 193, because the AKG C1000 was an electret microphone. I go into detail about this in my Problem Solving page.

For the recording stage, we used the knife to record the carrot chopping with different variations to accentuate power of the chop and the slicing sounds of the vegetable. We also recorded some of the sliced carrots falling into a metal bin for when they fall down the chute. These were then timed and edited into the timeline to match what's seen on screen. For the chute, I added reverb with a high and low cut which would accentuate the metallic echo of the carrots falling down. The result of these recordings and edits were very satisfying and I was confident they were of a professional quality that would immerse a watching audience.

We were able to use the remaining chopped carrots that hadn't been in the gross wastepaper bin for another part of the Foley recording: the large amount of bunnies feasting on a pile of carrots. For this I wanted a layering technique to accentuate the sheer number of rabbits munching away at carrots. So, we each took it in turns recording ourselves munching carrots slices into the microphone for a total of 5 layers of carrot munching: me, two tutors, and two classmates. The process was comedic but quite gross, being that I was the one monitoring the sound through the headphones (see images below). Nevertheless, the sheer variety of munching produced from that session produced a really convincing sound after they were all layered, positioned, and edited into the timeline. I was able to slowly introduce layers too as each bunny showed up in the scene. I even panned some of the layers slightly which gave the overall soundscape of the scene a wider feel too. The overall result was extremely high quality, and immersive (again, on track with my aims), and gross.

I also would like to mention, as it's somewhat relevant to this scene, I actually used the delay effect on the bunny footsteps as more and more enter to give the illusion that more and more bunnies were entering the scene. This worked very well to my surprise. It taught me how using tools creatively at in a DAW can have a positive effect on a project.

Lord of The Rings Highlights: Coconuts and Carriages

With the ambience and footsteps from my field recording already recorded and edited into my clip, the next important piece of Foley work for the Lord of the Rings clip was the horse and carriage. For this, I had already researched that hollowed coconuts provide a convincing sound effect when clapped together for horse hooves. So, I bought a coconut, drank and emptied the water inside (it doesn't taste phenomenal), and scooped out the white inside (which also doesn't taste great). I then washed, scrubbed, and dried it so it was ready for recording in my next studio session.

With the coconut prepared, I went into the studio and followed my usual process as mentioned before. I experimented with different ways to clap the coconuts together, with different amounts of force, different angles, and timings producing different effects. Ideally, I would've wanted to use a foley pit with the coconuts to simulate them trotting on a grass or dirt surface as seen on the screen but that was unavailable to me. Given more time, I think I would've been able to achieve this. When I had a motion with the coconuts that I was happy with, I recorded it paying careful attention to my timing into my project. I think what was achieved was good. It's still convincing and immersive in my opinion but again, would've been better if I had built a Foley pit.

Now that the horse had been sorted, I needed to record a Foley layer for the attached cart. For this, I had researched a methodology that used a combination of a desk chair and a pile of wood. I found a desk chair easily in the studio and gathered a bunch of wooden planks I from a storage room. I then piled up the planks and placed the desk chair on top. For this recording, I used a Shure SM-57, a dynamic microphone. This was for a few reasons. From my research, I knew that it was less sensitive and I was going to aim to produce quite a loud sound from the wood and chair to get as much detail into the action as I could. It was also a bit less responsive to bass tones so all the frequencies made by the wood wouldn't overwhelm the sound effect produced. Also, it was a lot easier to position for the action I was going to do.

With my materials set up I followed my usual Foley recording process but I didn't necessarily need phantom power this time. I practiced the action, gain staged, and recorded the sound produced into my timeline. I then used the scissors and fade tools then lined it up.

An additional layer was added after this too. I noticed one of the planks of wood was carpeted, so I also decided to try and simulate the sound of the wheels on the dirt by scraping a half a coconut shell along it. After recording this and putting it in my project, the result was ok. I was very low on time at this point and this effect was very spur of the moment. It being layered with the other two effects helped a lot but again, I think this could've been done better with use of a Foley pit for better immersion in the final product.

I additionally added pan and volume automation to these three sets of layered foley to indicate the approach of the horse and carriage and the different directions it heads across the screen during the clip. I think the result for this, was very satisfactory, providing a convincing, immersive soundscape for the audience.

Using Virtual Sounds and Instruments for Foley

This is more of a side note than anything but worth mentioning. In my project there's roughly 3 occasions in which I use synthesised sound effects for my Foley. This was done using different virtual instruments built into Logic and a MIDI roll. This was something I wanted to avoid in my project but circumstance ended up forcing my hand or I was feeling like experimenting. The effects made from these were pretty good though.

The first one I used was for the doorbell. This was the Tubular Bells instrument and I timed it to play E - C along with the doorbell in the MIDI roll. I lowered the velocity of the notes a fair bit here because they were very loud initially. I think this worked well considering I couldn't find a doorbell that either worked or sounded like the one seen in the clip. What was produced was received well in my projects peer feedback although the virtual instrument did distort in the first render (you can hear this in the feedback version). I did nothing to fix this. It somehow fixed itself. I have no idea how.
The second use was in the Wallace & Gromit clip. This one wasn't exactly necessary and probably would've belonged in the soundtrack. I quietly played (technically programmed this time) a little arpeggio on a xylophone on a MIDI roll when the rabbit appears out the teapot. Again just a extra little thing that wasn't needed but I thought it was cute so I decided to keep it as something to make the audience smile.
Almost immediately after the rabbit appears from the teapot, they jump out. For this I found a virtual instrument simply titled "Comedy SFX" which has a variety of comical sounds. I decided to use one of these comic sounds when the bunny jumps out. I actually don't think this was very good and should've done something different for a higher quality final product. Maybe if I had found a way to produce my own "bounce" sound effect? Again, I think I wanted to make the audience smile with this one. It's ok. I don't have much to say on it.

Finalising Practical Work and A Short Gallery of Foley Materials

I got through the majority of my scenes before I had to declare a cut off point because I was running out of time. I decided to finish the clip at the point when Gandalf's carriage is by the canola field. In all fairness, I thought at one point I wasn't going to get to the third clip at all so I'm happy that I got as much done with the time I had. But, I had to stop midway through clip 3 because I was out of time. So, in my final studio session I did my final checks. I went through and analysed my foley, editing, and effects to check and see that they were in time and accurate. I checked my automation and panning to check it was correct too. I also made sure the levels of each sound effect were about right, not being too loud or quiet, adjusting each effect and layer as needed. These small polishes helped the final quality of the final product so I could maintain the final aims of my project.

Unfortunately, I didn't have time to record the fixes I wanted to apply after receiving peer feedback from my first clip. These criticisms can be read in my Evaluation page but basically there were a few things I wanted to add such as rolling glass eyeballs in the drawer and sounds to the hat rack in the Toy Story 2 clip. I simply ran out of time and didn't get the chance to do this which isn't the best. Either, what I had was very high quality.

I bounced out my final project (cutting out the sections I didn't reach) into a .mov format which was then uploaded to YouTube and can be seen on the "Final Product" page.

To finish off, here's a short gallery of some of the materials I used for my project: