A lot of musicians layer field recordings with their music. I’ve learned a few techniques to get the layering about right.
Last Updated: 06-October-2026
Levels of the Field Recording Within the Mix
What I am talking about here is a long-form field recording placed behind the music in a track and the relative levels of each. This is not about short-term foley effects as those can be of varying levels.
This also assumes the basic cleanup of field recordings has been done: Master Field Recording: Post-Processing Techniques for Field Recordings
The Field Recording Level in a Music Track
When you get to the point of layering the field recording into your music track, one mystifying thing will likely be what the level and loudness of the recording should be in relation to the rest of the track.
Whether it is sparsely layered or a constant background sound, such as in a room or an open field at night, the idea is the same: the level should be about the same.
I’ve found that if I place a field recording in a layer behind my music, the background sound of the layer needs to be -20 to -30 LUFS. Mine usually come in around -26 LUFS, and that’s measured through the master and everything, not as an individual channel. I check this on my tracks by turning off the music and percussion tracks and just listening to the field recording through the master bus while watching my meter. I look at the short-term LUFS reading during the steady background sound and not the events. The events can be much higher. These are like cars passing or other loud noises that may occur periodically. For the steady background noise, I am looking at the short-term LUFS.
I always want the bed of field recording noise to sound natural and sit behind the music to give it space. If your goal is something different, then raise or lower the level accordingly. By all means, use your ears to judge it. This is only a recommendation.
The Field Recording – Do’s and Don’ts
See this post for things I tried and did not work: Master Field Recording: Techniques for Realism and Distinctive Style.
While my other post goes into some detail on what works and what doesn’t, it’s worth repeating it here in the context of the track because a few items will come up when working with it.
When finishing the field recording portion (not the electronic music or the whole track), I try to preserve the realism and spatial detail captured in the original recording. That is my main goal.
Binaural recordings depend on subtle left/right differences in timing, level, and high-frequency detail, so I avoid heavy processing that makes the sound wider, softer, flatter, or more “produced” at the expense of the original sense of place.
My workflow starts with restoration in iZotope RX: gain if needed, de-hum, careful low-frequency cleanup, light spectral denoise only when necessary, de-wind, and surgical removal of clicks, rustles, or handling noise. I try to remove technical distractions without removing the natural texture of the environment.
Equalization (EQ)
I am careful with EQ. Low frequencies give a field recording weight and scale, while higher frequencies provide much of the detail and directionality. Instead of cutting large parts of the spectrum, I use small, linked-stereo EQ adjustments only when needed. The key here is linked adjustments (not independent left, right, mid, side), as this helps preserve the binaural image.
Tape and Sampler Effects
I have also found that tape emulation, sampler emulation, saturation, widening, heavy compression, and obvious stereo effects tend to reduce realism in binaural recordings. These tools can be useful when transforming a recording into music, but I prefer to keep a clean, natural version of the original field recording first.
If I do use a tape effect, the tape hiss and other noise are not desirable in these types of recordings where a field recording is layered in with the rest of the track. This is because the hiss competes with the field recording and often obscures or muddies it. So, where warble and instability can give an impression of artistic nostalgia to the field recording, I employ these but leave the hiss and noise out.
Stereo Imaging
Definitely, imaging (a.k.a. stereo enhancers) can ruin a binaural recording, but within a layered track where the binaural recording underlies other instrumentation, I find it acceptable to have small amounts of image enhancements applied to the whole track – field recording and all.
Low end, for example, needs a bit of grounding and bringing it closer to mono is totally acceptable. Also, the low mids can always use some enhancement since they are right next to the lows and this provides needed contrast. High mids I leave alone and the highs I enhance slightly. This works for me. Try different schemes on your own recordings and listen for that breakdown of the binaural and adjust accordingly to avoid it.
What I find is this:
| Band (in Ozone Imager, above) | Approximate Range | Setting | Effect |
|---|---|---|---|
| 1 | Below ~100 Hz | Width about –31 | Pulls the deepest bass distinctly toward mono |
| 2 | ~100 to 300 Hz | Width about +17 | Restores/modestly increases spaciousness in lower ambiance and resonance |
| 3 | ~300 Hz to 8-9 kHz | Around 0 | Preserves the original binaural relationship through most of the localization-critical midrange |
| 4 | Above ~8-9 to 10 kHz | Width about +6 | Adds a very subtle lift to air, fine texture, and diffuse openness |
Note the contrast in the width between bands one and two above. This is a true enhancement that really adds audible spaciousness without ruining binaural cues.
Reverb
For headphone playback, I sometimes add a very small amount of convolution-style reverb with Chameleon or some other convolution reverb. I use around 12% wet so it is just enough to give some effect but not obviously. The intention is not to make the recording sound reverberant, but to make it feel slightly more external and less like it is occurring inside the listener’s head. The exact convolution impulse response will vary, but I have found one that works for me, and it is detailed in my other post on the subject.
You can find recommendations on what worked for me here: Master Field Recording: Techniques for Realism and Distinctive Style
The basic idea is simple: preserve the location, clean up what does not belong, and use processing only when it helps the listener feel more present in the recorded space.
Levels of the Complete Track With Field Recording Backing
The whole track with instruments and the field recording, when finished, needs a target loudness also. For me, this is typically not going to be the same as if it were a dance track or a traditional electronic music track where it needs to hit -14 LUFS integrated or something of this sort. No, this is different for me. I always want the track to be highly dynamic, but also listenable.
Study Others
I studied a few artists to see what they have done in this regard. One was Geir Jenssen (Biosphere) and his Senja album which is a collection of field recordings overlaid by electronic music. This is exactly the type of music I was shooting for in my creations, so it was a perfect choice for study. You may need to choose a different artist and find one that is producing something like you want to make. Here’s what I found in Biosphere’s The Senja Recordings album:
Notice how the maximum true peak (Max. TPL) is right around 0 dBTP, and the integrated loudness (LUFSi) ranges from -29.04 to -9.98 LUFS, while the dynamic range (DR (PMF)) is between 4 and 18.
This type of measuring of other artists’ works helps shape my own targets, and here are the targets I find work for me right now.
- Targets:
- Master most tracks between -15 and-18 LUFS integrated. -16 and -17 are my sweet spots where I’ll try to be most of the time.
- Let quiet, documentary-like tracks sit around -19 to -21 LUFS if that is where they feel convincingly real.
- Let the more musical or rhythmically developed tracks reach -15 to -16 LUFS, without forcing the quiet pieces up merely for numerical consistency.
- Individual tracks should be around DR 9-15 if possible. Some can be lower because drones and these types of tracks do not inherently have dynamic range.
- Aim for an album-level dynamic character around DR 11-13.
- Set my limiter for -0.1 dBTP but do not fret if my tracks do not reach that peak. If I were afraid of distortion, then I would set my limiter lower like at -1 or -1.5 dBTP, but I am not intending to hit the limiter hard, so -0.1 dBTP should be sufficient.
The overall idea here, for me, is to get the most dynamic range out of the track while keeping in mind that it cannot be too much where the overall integrated level is too low for good listenability. This is a balancing act.
How I judge this, after they are rendered, is by listening and watching meters and there are a few steps:
- First, play the track without the true peak level hitting the limiter. That is, if my limiter is set at -0.1 dBTP, which it is, then the highest peak in the playback or rendered file must be below that.
- Now, is the integrated level in my acceptable range? This range varies for me, so see the numbers above that I tentatively use. In general, I want the integrated level to be around -18 to -15 LUFS.
- Is the DR level 10 or higher? If so, this track has potential to be perfect as-is.
- If the integrated level is too low, I re-run/re-render the track with the limiter input turned up until the peaks just hit the limiter (-0.1 in my case).
- Now, is the DR level 10 or higher, and is the integrated level in my range of -18 to -15 LUFS? If so, then I am done here.
- If not, then I keep adjusting and tweaking until I get the right levels in both DR and integrated loudness.
- But what if the track has a low DR from the start? If this is the case, then I need to listen to the material and determine if that is just the way that the material is, or is there a processor on an individual channel or bus that needs to be adjusted or deleted? This is another iterative series of hunting and adjusting before starting over from step one here.
Adjust and Finish
Sometimes the initial DR level will be low, and this means I am either limiting too heavily in my limiter, or my track itself is not dynamic. For the latter one, it is not so easy to adjust, and means changing the dynamics throughout the mix (backing off saturation, changing EQ, reducing compression, etc.). But, quick wins can be had with the first one by just reducing the level.
I make adjustments to levels of my field recordings behind my track and the processing through my master to target these full-track levels and here is how my tracks measure up:
