Brief Summary

Recording Classical music introduces a different set of challenges to those faced by people recording more contemporary styles. Faithful capture and reproduction and a policy of do no harm are appropriate here. Allesandro Panella shares his advice.

Going Deeper

As sound engineers, our everyday working lives are often characterised by a series of technical and artistic challenges that allow (or force) us to use our knowledge and skills in new and creative ways, so as to be able to overcome the problems we are facing at that particular moment. 

Obviously, these challenges might be unwelcome, but if we look at them from a different perspective and we take them as opportunities, our job immediately becomes a profession that allow the highest level of freedom, which, in my opinion, is a key element in professional satisfaction. We can devise new solutions to specific needs or problems, we can experiment with an infinite amount of processing devices and plugins to go where no-one has gone before, we can also lie to ourselves and to the world when we place a microphone poorly, get an awful sound and then say that the result is exactly what we were looking for (always for artistic reasons!).

But there is ONE situation where we need to put our creativity aside and take a step back, to become neutral “ferrymen” of the artistic message being delivered by one or more artists, and that is the production of classical music.

You might not envisage yourself recording classical artists, but it might well happen one day or another. It can be anything from a solo artist, to a string quartet, a woodwind ensemble or a full-blown symphonic orchestra.

In this article we’ll will choose the “small ensemble” scenario. 

Mind-set

If you are new to this, the first and probably best advice I can give you (besides knowing your basics of sound engineering, of course…) is to try and make all the process as transparent and straightforward as possible, both in terms of sound and of technicalities, starting from your intention. Most classical musicians don’t even know what DAWs, compressors and EQs are, they focus on their instruments and on the music they are playing, because that is their job and they have to respect the great composers who wrote music of eternal beauty. 

And we also need to be as respectful both to the composers and the musicians, so we need to help them through the technical stuff, so that they can concentrate on delivering the performance. 

This attitude translates into the following: 

• Be prepared: the studio (or better, the hall where the recording will take place) should be 90% ready in advance; you have to know how the instruments behave and interact with one another and with the environment, and you need to know the music you are going to record (new compositions being recorded for the first time are obvious exceptions);

• Try not to use technical jargon; 

• Be ready to interpret any requests that might be formulated with wrong technical terms.  

• Only offer constructive critiques as regards the musicians’ performance, if you are asked for your opinion. 

• Always remember: transparency is the goal.

Recording

The first thing to have to take into account is the space in which the recording is taking place.

Classical music is more heavily dependent on the acoustics of the space the musicians are playing in than any other genre, as most of the compositions were conceived to be played in a specific type of space, and the final result is greatly influenced by the reverb and acoustic character of the room or hall we are using. As a general idea, we should consider the room as an “extra musician”, which can give a great contribution to the performance.

Because of this, the lack of good reverberation in a small environment such as a typical personal studio might cause difficulties to the artists, resulting in a flat and problematic performance. Using artificial reverberation gives excellent results in other styles but classical performers are likely to be uncomfortable in a dry acoustic and headphones are often unfamiliar and rarely welcome. To solve this situation, consider moving the recording session to a local concert hall or a church, depending on the music.

The next step consists in choosing the right gear. As we are aiming for transparency, we need to choose the most linear and uncoloured tools we have at our disposal, as classical musicians tend to feel a particular affinity to their personal sound, they might not “recognise” themselves if too much coloration comes into the equation. As an example, some years ago I had to record a pianist in the small hall of his apartment; at the time I didn’t own high-end microphones and converters and because of the limited budget, renting top level gear was not an option. Adding the fact that the acoustic response of the room was horrible which forced me to close-mike the piano, the final result was far too “pop-like” and that caused the pianist to complain that he could not recognise his beautiful touch. To be honest, he was a bit too indulgent to this regard, but he was the artist and the artist (being the client) is almost always right. In retrospect, either I should have not accepted the job, as I was not technically prepared to face that situation, or I should have advised the artist on the possibility that the result might be different from his expectations. 

Fortunately, most of the mid-level gear we can buy today is more than adequate to this regard. We just need to know our tools and choose the ones that can yield the highest possible linearity.

As an example, I would choose Telefunken M60 microphones with TK61 omni capsules, which show a sufficiently flat frequency response and a very good performance vs price ratio. As a preamp my choice would be either a multi-channel Focusrite ISA (they can be found at very good prices in the used marked) or rely on a mid-to-top level audio interface (Universal Audio Apollo X series, Antelope Audio, and the other usual suspects).

Moving to the best mic technique to use, we should decide according to the line-up of the ensemble and the room acoustics. Generally speaking, I try to avoid spot miking plus room mics, as I would have to recreate the panning and balance of the instruments at mixing, which can be quite cumbersome.

As a rule of thumb, a good stereo recording is always the best option for classical music, and we have plenty of possibilities to choose from. If the room sounds good and has adequate reverberation, and if I have three identical omni microphones at my disposal, I would choose a Decca tree; this solution can capture the natural balance created by the interplay between the musicians together with a good amount of reverberation, even to the point where no mixing is required.

Alternatively, we can choose a Blumlein setup, taking care of choosing the distance from the musicians so that the balance between direct sound and reverb is correct, or even an X-Y, ORTF or A-B technique, paired with a stereo ambient recording, which will require balancing the various sources during the mix.

As always, the key is “keeping it simple”, so we should choose the solution that is both technically adequate and less difficult to implement.

As regards session parameters, lots of pages have been written on the topic, so I will not try to summarise them here. My advice is to choose a format which gives enough headroom and information detail while not placing a burden on the recording system; a 48KHz/24-bit session should do fine in most situations. Obviously, you can go to 96KHz and beyond, but in my experience the advantages are questionable. 

The last (and probably most important) advice for the recording phase is: be patient. The average classical musician needs time to get accustomed to the new space and acoustics, to the microphones and all the stuff that comes into play when a recording is being made. Don’t rush them or create pressure, let them repeat a passage as many times as they feel necessary, take all the notes you need on the various takes, wait and you will be (hopefully) gifted with some beautiful music!

Editing

This phase can be stressful, depending on the results of the recording session.

There are many factors that can make the editing phase an easy job or a nightmare, for example:

• How many takes were recorded? 

• Does each take contain the full piece or were the takes only partial?

• In the latter case, how small are these fragments in terms of bars or sub-bars?

• Was the environment free from extraneaous noise?

• If not, are the noises sufficiently low to be easily removed?

The possibilities are countless, so we need to focus on a single goal: keep the musicality of the performance intact.

This might be an easy task if we have “full takes” of the performance, but if the musicians opted for a fragmented recording we need to be very careful with the choice of the good takes and we have to be very precise and “smooth” when making all the crossfades between them.

In terms of noise reduction, we have to be as “light” as possible, to avoid alterations of the sound of the ensemble. iZotope RX can be our best friend, both for broadband noise reduction and for detailed interventions via Spectral Repair. In general, however, we should try to obtain a noise-free recording and use noise reduction tools only when the audio material is not usable “as is”. In my opinion, however, a great performance with some noise is more desirable than a noise-free recording of something that sounds like a mere exercise.

Again, patience is our best ally in this phase. Take all the time you need to listen carefully to the takes, choose the best ones in terms of musicality and sound together with the musicians and use all the skill you have to obtain a final comp that sounds as a single performance.

Mixing

As usual, this stage can be anywhere from pretty straightforward to extremely problematic, the discriminating factor being usually the result of the “GIGO equation”: if Garbage came In, then Garbage will go Out. If we failed to obtain a good sound at recording, then we will have far fewer tools at our disposal to try and correct the material than we have in other genres, which are more tolerant of heavy processing. 

However, assuming we have good source material, we should always try to reproduce what was heard live while recording, in terms of panning and balance among the musicians and with the reverb, depending on the recording technique we used. 

As regards processing, we can use a bit of equalisation if we feel that our microphones coloured the sound too much, and I generally tend to operate only with subtractive EQ, mainly to attenuate any resonances that may have been captured. If I feel the need of more “presence” or incisiveness, I tend to try a very small amount of saturation, which can yield better results than boosting the high frequencies on an equaliser, while also creating some sort of global glue or cohesiveness to the material (again: just a touch of processing here!).

The most difficult situation is a session with multiple microphones and no ambience. In this case, we have to recreate the whole setting (balance, panning and reverberation) in our DAW; as regards panning, if you are not sure of the positioning of the various instruments in the stereo field, don’t be shy and ask for the musicians’ help, they know where each of them should be placed. 

Balancing the microphones can be tricky, as classical musicians share with their “modern” colleagues the tendency to suffer from the “more me illness”. No easy solution here, just don’t fear to point out that an instrument is too loud if the musicians’ requests become counter-productive. 

Speaking of balance, with a digital reverb (a convolution reverb is a possible candidate here), work your way up slowly, increasing the reverb volume in small steps until you reach a balance that both the musicians and you feel is correct for the music being played. Sometimes, we may need to use the reverb eq or HPF to eliminate some build-up in the low range, but that’s the only “extra processing” I’d typically use on the reverb channel.

Mastering

In my opinion, the mastering phase should be limited to minimal interventions, aimed at creating more cohesiveness in the audio material, and only if this is really necessary. No fancy stuff here (e.g.: don’t even think of cutting bass frequencies in the side, or using image wideners and so on). The only processing I would take into consideration is as follows: very small eq moves, maybe a touch of saturation (if I haven’t already used it while mixing) and super-gentle limiting. Why? Well, first of all the desired global timbre of the recording should be already there from the previous phases (so very limited corrections should be needed); second, dynamic range control should be left to the players and their musical abilities.

It may happen that the musicians ask you to compare the impact of their music and some modern productions. If you find yourselves in such a position, try to explain that “impact” is not what they should be looking for, which should usually be “emotions”. If they what more volume, well… they can turn the volume knob up!

Final considerations

This is my personal view on how I would manage a small classical ensemble. A full orchestra will need a different treatment (at least in the mixing and mastering phases). As always, these indications should be taken as a starting point to make your own decisions. Bear in mind that I started my career as a classical pianist, so I might be a little too much of a purist when it comes to recording classical music. And always remember, it’s the results that count, the listener doesn’t know how you got there.

Let us know what you think about this topic in the comments below.