Wolves. Chapter 4. Techniques and approaches.

The Goal

My goal was to record enough various isolated wolves material (no insects, birds or other environmental sounds) to make a comprehensive cutting edge quality library of wolves’ vocalisations useful for film sound, game audio and other media production.

Equipment.

I am not endorsed by any equipment manufacturer. I share my own experience and decisions I make to achieve my goal.

I took a lot of mics with me but practically used 3 different systems:

• Nevaton MC50 Quad large condenser in Rycote WS AG Windshield Kit. It’s a coincident quad mic with precisely matched cardioid capsules pointing at 4 directions with 90 degrees between them. Their frequency response specified by the manufacturer is from 5Hz to 35 kHz. The mic has internal heating and is very reliable in humid conditions.
• Sennheiser MKH8040 ORTF + Centre system in Cinela blimp (thanks to sound designer Alexander Volodin for giving me the system). Their specified frequency response is 30Hz-50kHz.
• Neumann KMR81i short shotgun in Rycote WS4 Windshield kit. It’s frequency response is 20Hz-20kHz.

  

My main recorder Zaxcom Fusion 12 was busy at film set at that moment, so I took my Metric Halo ULN-8 8-channel mic preamp and interface and a 2013 Retina MacBook Pro. May seem a not so mobile setup, but given that at that time the territory of the shelter was rather limited, and the wolves were kept in open-air enclosures roughly 20 x 10 metres, AC power was available on the territory, and with mic cables at least 10 metres long I could place the interface and laptop on a foldable table and sit there hidden in small pine trees with the mics placed on stands close to the enclosure fences. ULN-8 has fantastically clean preamps with precise stepped analog gain controls before ADC, that can be linked across groups of channels. There are no limiters, but the inputs can handle a signal up to +20dBu without analog distortion and the levels behaviour is very predictable. As long as you set the gain with enough headroom, you do not have unpredictable level spikes I’ve encountered with some other preamps due to distortion in the analog stage. I had my maximum peaks a few dB below 0 dBFS, but still was able to add up to 25 dB of gain in post for soft sounds without noticeable preamp or ADC noise. The only weak link appeared to be the generic switching power supply that is shipped with Metric Halo units – it sends a band of noise above 65kHz into the DC power feed and affects the corresponding frequency range of the recorded signal. This noise is below the noise specs of the unit, but still visible on the spectrogram and may be critical with low level signals and mics such as Sanken CO-100K. Luckily I had no useful signal above 65 kHz, so I applied a high-cut filter in post. Otherwise, I recommend using a better transformer-based and well-filtered power supply or a battery. The interface and laptop system successfully worked outdoors down to -10 Celsius (14 Fahrenheit), but if I closed the laptop and it went into sleep, it would’t wake up and had to be taken to a small warm cabin to heat up and come alive.

Techniques

Some colleagues consider using surround microphones excessive even for recording ambiences, and when it comes to effects it’s a complete no-no. Nevertheless, from the first recording day in the wolves’ shelter I was sure that surround was the way to go for recording collective howling. I settled in the central enclosure literally surrounded by small groups of wolves in their enclosures situated around mine. I was dreaming of film scenes in which we, being characters and spectators, are in a forest and wolves hiding around howl scarily. Or we are among the members of the pack, joining their chorus around us. I placed the Nevaton Quad in the central spot between all wolves, Sennheiser system closer to the alpha pair, Neumann closer to the older wolf Skif on the opposite side. I was immediately rewarded this evening with a series of chorus perfomances of the pack. Now, watching the animated feature film “Wolfwalkers” that uses a lot of my wolf recordings, I see that recording surround was an absolutely right decision. A lot of scenes are built the way that we are surrounded by howling wolves and these scenes feature my quad surround recordings.

From that night on I constantly recorded all the outputs of the Nevaton quad, but when placing it close to one group of wolves I oriented it as L C R with one cardiod capsule aiming at the centre of the sound stage.

This way I had the 180 degrees wide stage covered with 3 separate capsules of equal quality, aiming at their respective directions. As long as they are coincident, I could choose the best sounding capsule or mix 2 or even 3 of them for close-up but wider image. This is the way the quarrel of Sever and Hort is recorded and mastered and many other close recordings of the wolves.

The Sennheiser LCR system due to the fact that it’s not coincident, proved to be useful for more distant perspectives panned into L C R speakers, while Nevaton Quad works well both for close and distant recordings.

Neumann shotgun helped mostly with quiet sounds with narrow frequency range like breathing, chewing etc or for picking a distant but interesting voice in the wolves’ chorus.

Approach to the talent

Day schedule

In my experience with these wolves and even animals living in other shelters I visited , they have a certain schedule of activity that happens not only at night. Maybe it’s the natural biorhythm or they somehow adopted to the human activity. They become active in the afternoon, but you cant’t record anything useful until it gets completely dark at 7 PM because of birds singing even in cold November. Then you may record them eating and walking around. At 8 – 9 PM wolves collectively howl a lot till 10 pm and settle for an hour of rest. Then howl again for half an hour and rest for 2 hours. The old wolf Skif often sighs mournfully and grumbles during these rest periods like an old man speaking of his rough life. After a short period of activity around 2 AM wolves settle for a sleep till dawn with very sparse sounds in this period. It’s the coldest part of night and wolves prefer to sleep through it. On the contrary, every night exactly at 3 AM foxes in their enclosure got very noisy yelping, yapping as if some one of them is being killed, and rattling the metal roof of their house. Dawn came at 7:30 – 8 AM and when I witnessed this for the first time, I was amazed that every morning wolves sing their dawn chorus celebrating the sunrise! After that they eat and then howl again and get to sleep at 11 AM. This period of morning activity may be lucky for recording or not, depending on the birds activity, depending in turn on the weather.

Recording animals always includes many unpredictable factors that may work against you or in your favour, and the effect of these factors is not quite obvious. Of course, rain or snow will ruin a whole night of possible recording. Humid weather may seem bad for the mics but will suppress birds activity. Sometimes there are nights when all dogs in the nearby village get crazy and their barking turns into an all-night pulsing drone rendering your recordings useless.

Attended versus Unattended and the measure of intrusion

Most of the time during that week I was the only human in the shelter. Obviously, without my presence wolves feel and behave more naturally. Vlcak Grom howled and mourned so emotionally when me and my mics were away but kept silent for hours in presence of the mics even if I was away. I gave up and put only one mic in front of him for the last 2 hours of my last day at 9 AM just in case, and went away to the small cabin to drink tea and chat with the keeper of the wolves. To my surprise this turned out as 2 hours of clean in-your-face solo howls, growls and mourns.

On the other hand, it’s always interesting to know, what was happening, to be able to describe and attribute the vocalisations correctly. Besides, I really get adrenaline while recording, even if it’s bird songs, not even large carnivores. I can’t live without this hunter’s drive. So I kept the balance by sitting and monitoring in the distance, hiding in the small pine and spruce trees. Of course, the wolves knew that I was sitting there, but at least my presence was not so annoying. Wolves’ eyes reflect light and it’s an additional thrill seeng their eyes staring at you in the darkness.

Recording wolves eating seems obvious: set up your mics and give them their meal. But even if this meal is a cow’s head weighing 10 kg, a wolf will rapidly drag it to the furthest corner of the enclosure and eat it there far from your mics. I suppose next time I will place a spot mic exactly in this furthest corner. Without this possibility I invented a trick: I lit the corner with a spotlight and the wolf felt more comfortable in the shade close to my mics than in the lit corner.

I always feel negative about provoking an animal’s aggression for recording purpose. Aggression is the result of fear, an animal is being traumatised when provoked for aggression. Animals are so clever and emotional, they feel love, jealousy, care for the younger and weaker members of the pack. They didn’t choose to live in enclosures instead of wild forests. I find immoral disturbing them, I just observe and record their social life and donate some part of our income to the shelters. And even if someone has no moral barriers, it is merely dangerous to tease a big animal next to you, even separated by fences and bars, no fences are perfect.

During my first night I tried to chase events. Whenever an interesting activity started in one of the enclosures, I would pull my mics to that part. Often while I was relocating the mics, the activity stopped and something interesting started in the opposite corner. Then I understood that indeed, I should go with the flow, but rather than pull the mics now and again, I should spend more time watching and guessing what enclosure is going to be the main stage for the current night and prepare to capture that main show. You always have to be ready for everything, but have to choose the main stage and keep to it, sacrificing the smaller events. This method worked with the quarrel of Sever and Hort because from the beginning of that night Hort’s territory and its border with Sever looked like the centre of emotional excitement.

Studio work

I brought back about 60 hours of multichannel and multi-microphone recordings of wolves’ life. I decided not to include multiple mic recordings made at the same perspective into the library, but keep only the best options. When I hear voices of large animals within metres or even centimetres to them, I’m completely overwhelmed by the richness of their timbre. Infrasonic rumbling and very high frequency yelps and breaths spreading far into ultrasonic range. Even a cat’s purring has a lot of low frequencies when being listened to closely. I always struggle to preserve and convey as much as possible of this impression of a close contact with a wild creature. On the other hand, we had to make the final product as clean as possible. This goal implied certain choices:

• Considering the kit of mics I used, Nevaton MC50 proved to be the best for the task, with frequency range from 5 Hz and practically up to 45-50 kHz. Although it is not advertised as an ultrasonic mic, it outperforms Sennheiser MKH8040 due to better signal to noise ratio both in audible and ultrasonic ranges. This is the recording made with 3 microphone systems placed similarly. Download and listen. All rights reserved.

https://www.dropbox.com/sh/be75r7e7lq7gllc/AACI2sq5I636BatX3EfL9z2Xa?dl=0

• All recordings with even a hint of airplane noise in background went into trash. There is no way to remove planes and keep the broad frequency range of the recordings.
• Low-cut filter in post was used sparingly or not used at all to preserve the low frequencies of wolves’ voices.
• Due to the presence of AC power, some hum removal was necessary, but it was individually tuned for each mastered file, because even Absentia DX in Smooth mode can eat into long vocal notes.
• De-noising was used only for a few narrow-band sounds like breaths. I have any algorithm available, but all of them, while making the recording technically cleaner, make it sound duller and more distant. I prefer to keep the immediacy and agree to the slight movement of air a bit noticeable with soft sounds that I amplified during mastering. It will be inevitably masked by ambiences in the mix.
• Me and Natalya Dmitrieva have spent an enormous amount of time in Izotope RX doing spectral copy-pasting, de-clicking, de-rustling and other sorts of spectral surgery around the useful sounds, removing footsteps, movement noises etc. This is an inevitable part of animal sound library production.

All the naming and metadata work was done in Soundminer. Like a heavy stone falls off my shoulders when I load the files into Soundminer and see the final structure of the library being implemented. I feel special pleasure while I embed photos into the artwork field. And then there is the cover, designed by Natalya. It feels like an album cover from a vinyl or CD era, a cover as the essential part of an album’s artistic concept.

And then there is this thrill of releasing it to the public and waiting for feedback. I really need you feedback, my friends and colleagues.

And the last question:

Did I try to talk to the wolves? – Yes, sometimes I started asking them in Russian to howl or to produce some sound and then I thanked them. Did I try to howl myself to start them howling? Yes, I had seen Anastasia, the shelter founder, doing that, and tried myself. Not that I was pretending I am a wolf, I was just trying to communicate. When I listen back to my howls in the studio, I find them ridiculous, but at the end of the week in the shelter the wolves sometimes answered me. These wolves just spend time and communicate with humans. It seems to me that they learnt to understand our language much better than we understand theirs.

Buy the Wolves sound library. 10% of it’s revenue goes to the only shelter for wolves in Russia