Throughout the previous installments we’ve explored many of the individual facets of audio mixing. I thought it best to end this series with a real world look into the mixing process.
I’ve been fortunate over the last year to work on a project for Lianna and Elias Berlinger who are super talented musicians and songwriters from California. I’ve chosen two songs from this project to help illustrate several of the mixing techniques we’ve discussed in previous posts. We don’t have time here to outline every nuance so I’ll highlight some of the more interesting treatments.
by Lianna Berlinger
Examination: Let’s take a listen to a rough mix sent to me by the artist.
Prognosis: Even though this is not a finished mix I could tell this was a great song and the challenge for me would be to bring out the very best it had to offer.
Every engineer has their own approach to start a mix. It varies with each song but usually I focus first on the drums, bass and vocals in that order. Especially when live drums are used with multiple microphones.
Examination: Let’s listen to the raw main drum tracks.
Procedure: In this case I used the overheads to add some parallel compression. I added a good amount of compression (4:1 ratio) as well as some EQ. Then mixed that track back into the main drum group. Since I wanted more control over the kick and sub kick, I cut the lower range of the overhead track so I could control the kicks as a group separately. And because I knew I was going to add lots of lush long reverb and delay to the main drum tracks, I discarded the room tracks altogether. Less is more sometimes!
Let’s listen to the raw drums and then bring in the overheads a little at a time.
Examination: One other treatment of note is the snare cross-stick or “rim-click” as some refer to it. One of the things that make live drums sound so good is the bleed you get from every mic. It’s often a huge plus to the overall live sound but can also give you less control and make things sound muddy if overdone.
Procedure: For the snare, I decide to use a gate to isolate the snare from the other drums in the background. A gate only let’s through the sound that occurs above a certain threshold. Once you have a nice isolated version of a snare it makes it much easier to add effective delay and reverb.
Note: Something like this may seem like a small detail but I often tell people that these details add up in the final mix. Reducing unwanted sound and gaining the slightest bit of control over a track can add up to a better mix in the long run!
Examination: Let’s take a listen to the raw vocal track.
Diagnosis: The first thing I noticed was a great deal of sibilance. The exaggerated sound of the “Ssss” coming through. The vocals where recorded in California so I not sure where the issue originates. Could have been the choice of microphones or other factors. With the effects I have planned for later these high frequencies will become even more exaggerated. So this is a problem that needs addressing.
I also noticed a frequency that was slightly unpleasant to the overall vocal sound. Remember that not all frequencies are created equal. There is a range of frequencies called the “pain frequencies” located around 1Khz-4kHz. These frequencies can be literally painful to listen too. Our ears are much more sensitive to this range being as it’s a key range of the human voice that we learn to focus on since birth.
Procedure: So first up was to add a de-esser to reduce some of the sibilance. A de-esser is a specialized compressor which targets the upper frequencies.
Next, I added some compression and a bit of EQ boost around 220 Hz to bring out fullness and 6 kHz to make it a bit brighter.
I added a separate EQ just to calm that “pain” frequency issue with a dip around 1.5 kHz
If you listen carefully you’ll notice overall the level is much more controlled and less harsh which will make it react better to the delay effect ahead.
Prescription: For effects I thought of adding a long reverb but remembered a technique using stereo delay that worked wonderfully. Any unwanted sound in the raw track will only be exaggerated with an effect of this type. So I also put a de-esser on the delay aux track and cut some of the upper frequencies.
Examination: One of the nice instrument elements is the acoustic guitar through out the song. But they need a bit of life and a space of their own in the mix. In the beginning they are panned hard right and left with an added guitar element coming in later.
Diagnosis: The challenge here is the lower frequencies. They give the acoustics a rich sound but can muddy things up when mixed with the lower frequencies of the drums and bass.
Procedure: After a bit of experimenting I used a high-pass filter and cut everything below 90 Hz. Also a little bump of 2 db at 1kHz. I used a Renaissance Axx compressor with a moderate setting and an attack of 5 ms.
Looking Deeper: This is where some thought is needed to the overall emotion and impact of the song thus far. Like most songs, it can be repetitive. And the challenge becomes how to make the song build and stay interesting through out. Can be especially challenging in a slow ballad. There are two places the song builds dramatically. In the middle and again at the end. Once the song builds in the center section it returns to the same instrumentation for the most part. So I decide to add more instrumentation with the hopes of enhancing the impact.
Here is the build up without added instrumentation.
My goal was to make the build up even more dramatic as well as add a freshness to the verses after the build up. So I added a synth sweep to the build up. Compare this with the example above and notice how the notes of the sweep carry you into the next section much more effectively. Once into the next verse, I added piano to give a fresh new sound to the mix with notes sprinkled in various places as to not interfere with the vocal.
This brings us to the master buss. Most engineers add a bit of compression to the entire mix even if they intend to send off to be mastered to help the mix gel together.
Prescription: Since I was in charge of the final master, I decide at this point to add a multi-band compressor as well. This allows more control over several frequency ranges at once and I can pick and choose which areas to compress or boost to effect the overall sound of the mix. A good multi-band compressor handles the transitions from one frequency range to another very well and adds to the definition and stereo width of the mix.
Certainly has come a long way.
In the next half of this last installment we’ll delve into different song with several different challenges. Hope you’ll stay tuned!!
by Lianna Berlinger
On this song I did supply the majority of the instrumental sans drums and vocals.
Procedure: As with the other song I decided to discard the room mic tracks and use my own effects treatment. I also used the same technique of parallel compression. I EQed out everything in the low range below 80 Hz and used a gradual shelf on the high frequencies starting at 4.5 kHz and up.
Notes: Again some of these changes are subtle. Listen back and forth and you’ll hear the second one has a rounder sound with a little more brightness in the cymbals.
Diagnosis: Levels are erratic. Needs more warmth and also a bit more life in the upper range.
Procedure: Compression and EQ added. 4 db at 220 Hz for warmth. 4 db at 4.5 kHz to cut through.
GUITARS, BASS & MANDOLIN
Diagnosis: Needs more impact as well as separation in the mix.
Procedure: Compression to make the sound a bit fatter. Panning and stereo delay to add width as well as EQ and effects to place it better in the overall mix.
Post-Op: I was at this point ready to get the first mix together.
Analysis: Though pleased with the mix I felt, like “Lullaby”, maybe it could have more impact in the chorus. So again I put on my arranger hat and go to work. Adding several layers of sampled strings. A base layer mostly in the mid to low range. A separate high part and a sampled “glissando” to really bring home chorus.
Diagnosis: Strings are a very dense and complex sound. Can mask the vocals as well as all the other instruments.
Procedure: To make room for the vocal I dipped the EQ at 3 kHz as well as cut the lows from 64 Hz down to clear room for bass and kick. To give them more stereo width I applied a stereo delay with one side set to 14ms delay and panned them hard left and right.
As you can see from both songs, mixing usually involves more than a few adjustments. It takes a good ear, experience with different techniques and a knowledge of arranging and how sounds work together for the greater good.
For this song, I like both versions for different reasons. The added string does increase the impact and gives it more of a pop sound. But I still like the more acoustic rock feel without the strings. So I will leave the final decision up to the artist.
Conclusion: That completes this series. I hope you’ve found it helpful and not only given you ideas for your own mixes but a greater appreciation for the people behind the scenes working to make your songs sound their very best!
Happy music making!