From Cable Basher To In-Demand Dialogue Editor: We Talk To Lucy Mitchell
- August 6, 2024
Growing up with a father who worked in the world of outside broadcasting and music video production, and a mother in Choreography for both stage and tv, Lucy Mitchell’s path into audio engineering was influenced early on by the fascinating blend of music and television. From her childhood experiences cable bashing at high-profile events like the Brit Awards and the BAFTAs to her work experiences across different facets of post production, Lucy’s career was shaped by a blend of practical exposure and academic pursuits. Despite early uncertainty about her professional direction, initially considering a career in business or finance, it was her hands-on involvement in sound editing during work experience placements at various companies in the holidays of her classical music degree that solidified her passion for the field.
Now with 150 IMDB credits to her name including Derry Girls, Top Gear, Death in Paradise, Boiling Point, EastEnders and The First Lady to name a few, Lucy is among the UK’s most in-demand Dialogue Editors. Here, we pick up at the turning point of Lucy’s career, the challenges of post production work and how Accentize plugins have helped to overcome them.
What was the tipping point at which you went from effectively being a runner to having TV credits?
I felt I had to work really hard to get to a base level of understanding. Most kids now come to the runner job having mixed films in 5.1 at university and I had no idea what any of that meant. I felt a huge pressure on my shoulders especially because my runner job was off the back of work experience I got through my dad, (even though most people knew someone at the time to get their interview). I felt like I had to work really hard to move up because I felt that I didn’t know as much as everyone else, so the competition was strong. I think in a way that worked in my favour. I became quite an important person when I was in the machine room and they created this Head of Machine Room role for me. I think it’s because I just naturally took control a little bit or just, as much as I hate the term, I got shit done. In this role, if stuff went wrong in the machine room it fell on my shoulders, so it was my prerogative to make sure that everyone was doing things well. It gave me a really good relationship with the post producers, because I was the person they’d talk to about things that were going wrong.
At the time ENVY were mostly doing documentaries – they did a fair amount of science programs which had CGI sequences to explain how a machine works for example which required more sound designy bits. The way that it worked at the time was someone would do the sound design, and then the mixer would essentially have that all on one fader, and they would just mix the rest of the music, voice over etc around it. So, they wouldn’t necessarily mix the sound design bits, maybe do a few tweaks if needed; someone else would do that and they would put that into their mix. At the time there was a very good friend of mine called Mary, who was mixing more but she had always been the person who always did those creative things before. They were like, ‘Mary’s mixing more and we need someone to be the sound design person’ so they asked me to do it while I was in the machine room, which surprised me because there were tracklayers who they weren’t asking. I don’t think it was that they weren’t good enough, the guys were great, I think it was maybe because of the relationship that I had with the producers – perhaps they wanted to give me a chance? I don’t know what the reason was. I was basically in the machine room doing the Head of the Machine thing and then going and doing the sound design bits as well. I started to do voiceovers not long after, while still in the machine room. Once I was promoted to Tracklayer and left the machine room, I was getting requested for voiceovers from clients, repeat clients, as well as for the sound design stuff. It gave me a bit of confidence that I didn’t have before because I was so intimidated by all the stuff the boys could do with all their prior knowledge.
I just started to do really well in that and I ended up doing certain projects that were always my projects, and that was really nice for me. I was a Broadcast Hotshot in 2013 (a Top 30 under 30 list in Broadcast Magazine) too. However, I was starting to realise that, because I didn’t really watch documentaries, the only reason I knew how they sounded was really from working on them. I mostly watch high end drama so I wanted to work on that because I had much more of an understanding of the sound of it – it’s what I consumed myself and I enjoyed. My issue at the time was, ‘how do I do that?’ I loved working at ENVY, and I was in a really good trajectory there. I was starting to pre-mix stuff, and I knew that if I stayed I’d be a mixer pretty soon. I had a really good relationship with the post producers, but I didn’t know what to do.
It came to a head when my mum was ill and I needed to be flexible to work at home. Although they were incredibly supportive as an employer, allowing me to swap shifts and work different hours, I started thinking about my career path. To cut a long story short, during a trip to Australia I started to think that I could do the freelance thing from home (I looked into working there and was told most sound editors are not staff over there. Even more so than England at the time). So, I emailed companies back in the UK and, I think because of my work at ENVY, I got a lot of interest and managed to secure meetings/coffees with pretty much everyone I emailed. I think what was very good for me going freelance, was having worked in somewhere like ENVY first, along with the amazing projects that I had worked on whilst there. Obviously, I couldn’t have gone freelance at the beginning of my career anyway because I didn’t know how to use Pro Tools or even what a WAV was!
Being freelance gives me a different feeling about my work than being a staff member. There is a sense of achievement when on a project when you’re a freelancer because people have specifically chosen to work with you. Even more so when you’re hired there again. They’ve offered you a job because they think you’re good, not just because you work at a company they want to work with.
Thinking about the challenges of your job, Boiling Point for example, for which you helped work on the dialogue, what kind of challenges did you face?
Boiling Point was an unusual one for me because I wasn’t the main Dialogue Editor, I was helping out their brilliant editing team James Drake and Oscar Bloomfield-Crowe. Typically in my job, if there’s lots of movement or someone’s chatting and they’re knocking things and all that sort of stuff, you want to reduce or remove it so you’re not distracted from the dialogue, or ADR it so the mixer has the flexibility to make the movements (replaced with sfx and foley) as loud or as quiet as they like under the dialogue. On the project that I’m working on now for example, one of the things I had to clear up with my mixer was ‘what’s your level here? Do you want me to ADR it all if it’s got any noise on it?’ because obviously as soon as you turn up someone’s microphone and they’ve got movement on it you’re making the movement louder, right? So there needs to be a limit of how much they’re happy to have on there or whether they want it completely clean, in which case I need to ADR the whole thing. Whereas on Boiling Point, obviously they’re going to keep a lot of the noise in, that’s part of the chaos of it, however, we also still need to hear what they’re saying. So, we often had to reduce knocks and movement rather than completely remove or replace.
The guys had mostly marked up or recorded the ADR before I’d worked on it so I was just doing my dialogue edit and then editing the ADR into it, and then reducing bangs and crashes where necessary. It was an interesting job because I was like, ‘I don’t know whether this needs ADRing or not, are you happy with the amount of knocks and bangs there are?’ because obviously it was so busy in the kitchen. The end result from James, Oscar and Jules Woods (re-recording mixer) was brilliant, the balance of noise, chaos and dialogue clarity was spot on – they were actually nominated for the sound BAFTA. It was really interesting to work on, I really enjoyed it. It was that kind of balance of how much noise is too much noise?
A lot of mixers opt for less noise reduction to keep the integrity of the voice and I think that's something that dxRevive does better than most.
Talking about noise, what are the general things that end up on your desk where you think, ‘Oh shit, I've got to figure this one out today’?
Luckily the projects I work on these days have good production sound teams, so usually things I need to approach with noise reduction are mostly out of their control. Clothes rustling is often a problem, and I have actually been using dxRevive for that a lot. Often I use blanket noise reduction for “noisy” scenes, when it’s just background noise that’s a problem, but because what I tend to do a lot with Accentize dxRevive is the four frequency band setting, I can go ‘it’s really rustly’, and can lower just the higher end, or high mids, and really target the specific noise and keep the resonance of the voice and that’s really helpful. So I’ve been using it a lot for that. I find that helps for reducing a lot of ADR.
External scenes are always tricky because every single take will probably sound really different in the background. The thing I’m doing at the moment, for example, is by water in one of the scenes and there’s a lot of wind and a lot of waves and some of the waves are really low end and some of them are high end and the movement of them is a problem. Obviously, if there’s movement, it’s harder to reduce noise, which is where the dxRevive multiband option helped. I’m more likely to bring up queries of noise reduction with external scenes just because the noises are so different. So, I don’t know if everyone else does this, but I do lots of really long transitions if I can, so that if the mixers are up against it and they don’t have a lot of time to EQ everything, or to match the sound of each take exactly, it leaves the transitions not as obvious. So, when it comes to noise reduction that helps because you don’t want to make everything else outside completely clean, that’s weird.
Someone did that to me once when I was mixing a few years ago – they had done so much noise reduction outside and it was weird because the actors were standing on a corner of a busy London street and suddenly there was no ambient noise! The voices also sounded dull because so much noise had to be removed to achieve what they had given me. I would rather have realistic traffic, a bit noisier, but with realistic voice, because that’s the thing; reducing the noise is affecting the voice, so a lot of mixers opt for less noise reduction to keep the integrity of the voice and I think that’s something that dxRevive does better than most.
Do you find the algorithms in dxRevive help you with that one? I'm guessing you probably lean towards the more natural algorithms than the more studio ones?
Yeah, and actually today, for the first time in ages, I switched them. I usually do Natural or Retain, but I switched this morning to “Studio” because there’s a scene where this woman is shutting her eyes and she’s whispering. It will be over music, which means because she’s whispering, they’ll need to push her levels up, which is turn means pushing up the room noise. At recorded volume the room wasn’t an issue but when the whisper is made loud, you suddenly hear it. I don’t really want to keep the background room hiss, I just want to hear her, so I actually changed the algorithm to “Studio” to achieve a closer, cleaner, more intimate version. I didn’t want the natural sound this time.
On another project I worked on, there was this one scene where we walked into a restaurant which had one of those open the kitchens at the back. The kitchen wasn’t open, they were prepping food, but it was quite loud because of extractor fans etc and they asked if I could work on it to reduce the noise because although realistic, the main conversation at the front of the restaurant was the important focus. I managed to do it so the voice didn’t sound overly processed, and pretty clean but they came back and said that although the voice itself sounded fine, it had lost all the life and energy from behind, which I totally agreed with. I said that I could keep the life in, but it’s going to be louder in the background and do they mind – again it’s asking what the mixer/supervising sound editor’s preference is in terms of how much noise is too much noise. Everyone is different.
The difficulty was the characters weren’t sitting next to the kitchen, they were the opposite end of the room. If they had been closer, it would make more sense for the background noise to be loud but they were quite far away, so came down to mix preference. It was at that point that I tried dxRevive – I redid the scene and they were happier. It is important to note that dxRevive is what I used for my pass as the dialogue editor – once the mixer works on it, they may do additional passes with their own tools. It’s a collaborative process, but dxRevive has been a really great tool for me to use as my main plugin to add to the overall process.
A lot of things can reduce a similar amount of noise, but they often ruin the dialogue more or it takes lots of different plugins combined to achieve a similar result which is also more time consuming.
You’ve obviously you've used a ton of noise reduction systems in your life but what did you think the first time you pushed that knob up in the middle of the dxRevive GUI?
It was exciting! At first I just did a blanket full denoise rather than multiband, and used some poorly recorded stuff I had on my computer. A lot of things can reduce a similar amount of noise, but they often ruin the dialogue more or it takes lots of different plugins combined to achieve a similar result which is also more time consuming. I think for some people with sound effects or for other things that they do noise reduction on, it’s not as obvious when you have artifacts or if you’ve gone that little bit too far, but for dialogue, it is so obvious.
If it’s a super noisy slate and you put the plugin to 100%, it’s obviously going to sound processed, but it was just really refreshing to hear something that sounded not so overly processed at a higher percentage (I probably wouldn’t ever use 100% anyway!).
The UI makes it pretty quick and simple to use. It’s the restoration side of it that helps I think, because a lot of plugins can reduce the noise itself quite well, but it messes up the voice too much. Some people don’t notice, but any audio person does, and it just doesn’t sound right. That scene I mentioned earlier sounded much nicer second time round than what I had done originally where it had taken away too much reverb and the space had lost its life.
Have you used Chameleon?
Yes, interesting you say Chameleon, because I’ve seen a lot of chat about it recently, and a job I did recently introduced me to it – it seems to work really well at copying the reverb of a clip. The dialogue editors working on that were using Chameleon for end of sentences often, if people were cut off, and on bits where, because it was a wide shot, the boom was really far away. So, they were getting the clip mic and then using Chameleon underneath as well, creating a sort of “wet” reverb version to mix the clip mic with.
That was something I hadn’t considered before in my editing, giving a wet and dry version. On a different job that I did a few years ago, there was a scene where it was outside and I did some noise reduction on it (I didn’t have Accentize at that time), I didn’t want to make it really clean because it was going to ruin the voice. What was interesting is that the mixer was blitzing it to essentially make it super, super clean, that you probably wouldn’t want to play on its own and then essentially mixing that in with the original or my version at a quite high level; again mixing a wet and dry version, or in this case a clean and noisy version. I usually give one track per mic as my final “version” but have started to do this sometimes if I’m unsure on the level of noise reduction wanted. As an editor, if you’re entrusted with the noise reduction, you’re often going to perhaps do it less than if you were mixing it when you know how you can make it work overall and also I would hate to give to a mixer something that’s super, super processed. There can be a fine line between clean and over-processed.
Sometimes, if you're in a really tricky scene, it does help just to know the context because I don’t want to waste my time trying to make it super, super clean if it doesn't need to be.
That must be really disconcerting having to do dialogues without knowing the context of what it's going to sit in at the end of the process.
You obviously have guide sfx and music but you have no idea what’s staying or going, or how the other sound team are changing things from the original offline guide. I actually messaged my supervisor on the job I’m doing now, asking for a bounce of the sound effects of this ep, even just as a WIP, because there is one scene I just need to check if spending ages cleaning the dialogue is integral (when you have tight schedules, you learn very quickly how to prioritise!). If they’re inside and it’s really quiet in the sfx, I’m going, “oh, this still sounds a bit hissy or a bit buzzy”, or, “I can get rid of the hiss, but it’ll make their voices sound a bit low end” or whatever, but if there’s going to be something going on in the background – traffic or there’s music playing on a radio or something, I don’t need to worry about cleaning it up as much if I’m low on time. And that’s the thing with experience, with dialogue editing you go ‘this is what I’m listening to in isolation, it needs to be as clean as possible, it needs to sound as good as possible’ and yes that is, to an extent, correct, and you certainly don’t want to hand it to a mixer going “I didn’t really do much noise reduction on this because I’m assuming it’s going to be covered in music”, because what if they decide to cut a part of the music cue? But, sound effects wise, at least on what I’m working on now, it’s not super stylised, only in some obvious sections and I highly doubt they’ll have a scene where they’re suddenly going to drop everything and just have dialogue, so having the sfx to hand can really help. If you’ve got a good team with good communication, you can ask, so you can check the dialogue in context, rather than second guessing.
Ideally, you’d make it sound brilliant as a standalone track so if they wanted to just drop everything and have the dialogue, it should sound good. But sometimes, if you’re in a really tricky scene, it does help just to know the context because I don’t want to waste my time trying to make it super, super clean if it doesn’t need to be, and I can spend my time making other bits sound better!
You've worked on some great stuff over the years but what one piece of advice would you now give to 16-year-old Lucy?
I think be more confident in your abilities and ask for more feedback. Unfortunately, especially as a freelancer, you often only get told about your work if you’ve done something really badly or wrong. Yes, you mostly only learn from your “mistakes”, but I was always so frightened of being told I wasn’t any good that I never asked; and didn’t really know if my way was good, or ok, or they were making lots of tweaks that weren’t considered big enough problems to tell me about.
Also, at the beginning, I had a great opportunity off the back of my work experience, with a great company, and I was never going to turn it down; I would have been mad to do so. But realistically, had I thought about it and gone, ‘I really want to do drama’, maybe I would have applied for a job at a post house like Boom or Molinare who do more drama. It took me a while to get to do those types of jobs because it’s not what I had in my credits. It’s that age old thing of they won’t give you drama because you don’t have drama experience, so it took me longer than I wanted. Had I started in drama, then I’d have more confidence in general because I’d know exactly how things should be in my editing work.
