This article will explain what is audio mastering, its goals, the process and its evolution: stem mastering and online mastering.

Why mastering is important

Audio mastering is the last, fundamental stage to correct and improve the sound of a track (or an album) before its publication.

Equalizers and compressors of the audio mastering studio

Let’s see which are the main functions of the audio mastering stage:

  1. First of all, with a good mastering service, it is possible to obtain a better audio quality. It is important to underline that audio quality is a very subjective thing! Audio mastering is art, not science. The mastering engineer will master the tracks following his own taste and his own style. He will put his “sonic signature” on the sound, like a painter puts his style on the paintings
  2. The second goal is loudness. If the client asks for an increased output level, the mastering engineer will try to achieve it. At the same time, he will try to minimize the artifacts (distortion, pumping, transients loss) associated with an high output level
  3. The mastering stage is necessary to ensure that the mastered tracks will sound good on any playback system: professional and consumer speakers, headphones,  car audio systems and so on.  This is possible only if the tracks are mastered with expertise and proficiency, using a linear playback system in an acoustically accurate room.
  4. Audio mastering is fundamental also to give coherence to an entire album. In fact, only with a good mastering service all the tracks of an album will be consistent in terms of tonal characteristics, output level, micro and macro dynamics.
  5. Finally, audio mastering is necessary to ensure that the mastered tracks will be 100% compatible with the distribution medium: CD, vinyl, audio files for iTunes, Spotify, Soundcloud or other online channels. In fact, every distribution channel has its own technical requisites; the mastering service has to “tailor” the tracks for the expected distribution channels. For example, a track mastered for CD could not be suitable for vinyl or iTunes.

It is not possible to obtain these results with automated services. They can only be obtained by a specialized mastering studio, with a good and focused experience on the field, a precise and linear monitoring system, an acoustically accurate room and good analog and / or digital tools.

Following there is an example of a track mastered by this studio.

Original mix:

Mastered mix:

 The audio mastering process

Generally the traditional audio mastering process (stereo mastering) follows these steps:
Three steps of the mastering process

  1. Conversion of the track in the digital format. If the mastering engineer receives the track in an analog format (for example on analog tape), it is necessary to convert the track in the digital format before mastering it. For this purpose, it will be necessary to use an high quality AD converter to record the track on the hard disk. This will permit to not loose the detail, stereo separation, deepness and impact of the original sound.
  2. Conversion of the track in 24/32 bit. Mastering studios request high resolution audio files, with at least 24 bit depth (32 bit floating point is even better), and at least 44.1 Khz sample rate. If the mastering engineer receives a 16 bit track, he will convert it to 24 or 32 bit. This will minimize the data loss during the mastering stage. Upsampling, which is the conversion to an higher sample rate (i.e. 88,2 Khz), is not strictly necessary. Some mastering studios do it, others don’t. It is not necessary because the modern digital plugins that generates higher harmonics (like some limiters and clippers) have an internal upsampling algorithm: this will reduce or completely avoid any aliasing effect.
  3. Track listening. It may appear trivial, but the first thing to do is listening to the track without any intervention, from start to end. This will allow the mastering engineer to understand which are the eventual unbalances and which will be the necessary interventions to do for improving the sound quality. This stage has to be made with the speakers.
  4. Restoring.  In this stage the mastering engineer will reduce the sporadic noises (like pops, sibilance…) and the broadband noises too (like hiss and hum on 50/60 Hz). This stage is not mandatory, and it is necessary only if the noises are too much strong. It is important to do this task using the headphones, because headphones are better than speakers to spot noises and hiss.
  5. Sound improvement and loudness.  This is the proper audio mastering stage. This step comprehends all the interventions (equalization, compression, limiting…) necessary to improve the sound and to obtain the desired output level. During this step it is necessary to use the speakers, because they produce a better perception of the “big picture”, the overall sound. I have covered this aspect in this article about the differences between speakers and headphones. The mastering engineer can realize the interventions with analog, digital or hybrid gear. This choice will determine the mastering typology (analog or digital mastering). There are different techniques that can be used during this stage:
    • Equalization, to achieve frequencies balance and to obtain a certain “color” of the sound
    • Transients reduction. This can be accomplished by using mono band or multi band compression, saturation (soft clipping), or limiting
    • Macro dynamics reduction, with mono band compression or automation
    • Stereo image widening, achievable using mid side equalizers, compressors or simple gain controls
    • Output level increase, using limiting or soft and hard clipping
  6. Corrections of the initial and final parts of the track. Often the mixed tracks are not optimized in the initial and final sections. For example, there may be too much silence in the initial or final parts, or an abrupt ending with a too much short fade out. In these cases, the mastering engineer has to correct these sections by regulating the silence and possibly by applying a fade in or a fade out
  7. Final track conversion. It is time now to convert the track in the audio format compatible with the distribution channel:
    • WAV 44,1 Khz / 16 bit for the CD format e for various online channels (Spotify, Beatport, Soundcloud…)
    • WAV 48 Khz / 24 bit for the DVD format
    • WAV 96 Khz / 24 bit for iTunes, vinyl, high resolution channels
    • DDP for the CD duplication
  8. Generation of other versions with lower resolution. Sometimes clients ask also for FLAC versions or compressed versions (MP3, AAC…) of the tracks. In these cases, the mastering studio will generate these versions from the lossless WAV files
  9. Metadata addition. The final step consists of the metadata addition (track title, track artist, album title, album artist, UPC/EAN code, ISRC code…). Today it is possible to add the metadata not only for the FLAC or compressed versions (MP3, AAC…), but also for the WAV files

Stem mastering

In the last years stem mastering has become a very common practice, but what is it? Let’s give two definitions:

  • Stereo Mastering (also known as “standard mastering”). This process has been described in the previous paragraph. It is the processing of the entire mixdown, comprehensive of every track. It is done to improve the audio quality and, often, to raise the perceived output level
  • Stem Mastering. This consists of three steps:
    1. Processing of groups (“stems”) of single tracks. This process can be simple (setting only the output level) or complex (in this case the single stems are also equalized, compressed and so on)
    2. Rendering of the mixdown
    3. Standard mastering of the mixdown

Stem mastering

So, stem mastering is basically a compromise between mixing and mastering. It is a mixing of stems (and not singular instruments), followed by a standard mastering of the mixdown. Stems can be of various typologies. Here are some frequent cases, but there are many different examples too:

  • All the instruments in one stem, all the voices in the second one
  • Main voice, other voices, instruments
  • Main voice, other voices, rhythmic section (bass, drums, percussions), other instruments
  • Main voice, other voices, rhythmic section, soloist instruments, other instruments

The complexity here is proportional to the number of stems, but depends also on their sound. If stems have already been properly  processed (equalized, compressed) in the mixing stage, then the mastering engineer could simply set their output level before rendering the mixdown. On the other hand, if stems have not been properly processed in the mixing stage, then they should be equalized and compressed in the mastering stage, thus increasing the complexity of the process.

All these aspects determine a higher price for stem mastering compared to the standard solution. Which are the benefits?

  • The main advantage is that you can process separately the voice from the instruments. In the stereo mastering process, the equalization impacts both on the voice and on the instruments. It can bring balance in the voice and unbalance in the instruments, or conversely
  • In a similar way, you can process the sound with less compromises. For example, it is possible to smooth out the mid high frequencies of the guitars without losing the presence of the trumpets; it is possible to increase the rhythmic section low frequencies without mudding the sound of the other instruments
  • All these aspects allow the mastering engineer to apply minor adjustments in the final stereo mastering of the mixdown

On the other hand, stem mastering is a longer and more expensive process than standard mastering.

From traditional mastering to online mastering

Mastering is progressively moving from a traditional model, made by attended sessions with the presence of both the mastering engineer and the client / producer, to an online mastering service. In this case, the artist sends online his tracks to the mastering studio. Then, the mastering engeneer works alone on the project and gives back to the artist the mastered tracks (or the DDP for the plant).

Sending audio files for online mastering

In the online mastering service, the client can communicate to the mastering studio his preferences. He can describe with words the sound and the output level to obtain, or he can suggest some reference tracks. These ones are necessary only if the client wants to obtain a particular sound in a very accurate way. If this is not the case, then it’s better to let the mastering engineer decide in autonomy the sound to reach. In fact, the mastering engineer experience often allows him to obtain better results when acting in autonomy, instead of following the strict guide imposed by some reference tracks.

The online mastering process certainly satisfies some practical and important needs. First of all, the client hasn’t to spend time and money for reaching the mastering studio. Also the mastering engineer has his advantages; for example, he is more free to decide when to schedule the mastering session.