17 min read
Humans have been interested in creating and compiling music for a long time. The music-making process includes several steps: recording, mixing, and then the final step of the equation, mastering. In recent years, we have been able to watch AI, Artificial Intelligence, infiltrate nearly every part of music-making. This includes parts of the recording process such as generating original drum parts, writing melodies to overall being able to create pieces of music that are extremely difficult to distinguish from those made by humans. Artificial Intelligence has now applied and integrated itself into the music mastering process and in doing, raised crucial questions about the need for humans in specialized areas throughout the music-making process.
Let’s talk about the Mastering process itself. Mastering is considered the final step in the post-production process of the music audio. Its purpose is to balance out the elements of the songs and fine-tune them into a consistent piece so that it sounds the same regardless of what platform it is being played on, for example, Spotify, CD, or iTunes. In other terms, the goal of mastering is to make the listening experience more cohesive from song to song. Without the step of mastering, the song sounds punchy and quieter. With a firm understanding of the goal, a mastering engineer will understand the type of sound that is trying to be achieved and will consequently help you achieve it. According to mastering engineer Ian Cooper, mastering is “a bit like photography – you can make the sky bluer, the greens greener.”
With all the revolutionary dynamics Mastering brings, it can also be expensive at the same time. Depending on the mastering engineer’s experience, it can cost from a few hundred to tens of thousands of dollars for a single track. These prices break the affordability for most indie artists and bedroom producers.
Fortunately, over the past few years, there have been automated options that have been introduced that promise bedroom/amateur artists access to professional-sounding mastering sounds without the costs associated with professional human engineers. This can be achieved in many ways. In some instances, deep learning networks are used, which analyze the input data, while others utilize a carefully crafted signal chain designed by humans which are deployed as software. Regardless of how these methods operate, the goal remains to master audio with a mere couple of clicks.
One of the most popular services for this purpose is Landar, which is hosted as a web service. One can simply upload the desired song onto the web service in order to get it mastered. Landr’s algorithm analyses this song and then chooses between three options based on how strongly one needs these effects applied, and then finally exports the results accordingly. The downside to this is that the effects are not all that flexible. For instance, if one is not satisfied with the mastered file that Landr gives you, there is not much that can be done about it in terms of finessing the sound the same way one would with a mastering engineer. The industry has mixed reviews of this as Landr: ArsTechnica, in a 2016 review, called Landr’s auto mastering an “auto-turd” whereas others in the industry claim that Landr does the job. It must be noted that Landr’s algorithm is improving with each song that is uploaded onto the platform. CEO Pascal Pilon, in an interview with The Verge, stated that “in 2017, we ran a series of blind tests with major labels and professional mastering engineers and LANDR was actually picked over some of the world’s best mastering houses.”
There has been a concern in the music industry that with the presence of Artificial Intelligence mastering services, the need for human engineers will eventually be completely eliminated. London engineer Streaky makes a remarkable analogy to mediate this dilemma by saying that it is similar to buying an off-the-rack suit. Someone who really cares about the quality of the suit will have a bespoke item made for him, but for the majority of people, the cheaper options make more sense.
Izotope, a software company, has taken approached Artificial Intelligence with a more educational lens. Izotope already has creates a popular line of plug-ins by the name ‘Ozone’ and for 2017, they added in an intelligent ‘Master Assistant’. Contrary to popular belief, the assistant does not do all the work. Instead, it gives a starting point that can be tweaked according to one’s liking. Consequently, producers can make informed decisions based on the choices made by the artificial intelligence model. An iZotope representative told The Verge in an interview that “It has nothing to do with competing with humans” and “For the fearful professionals out there, assistive technology minimizes time-consuming cleanup work so that they can hone in on the creative side of things.”
Adam Love, MajorDecibel founder, states “It is not a replacement for mastering done by a mastering engineer. Mastering engineers can provide feedback to the artist about their mix, hone in on a particular style, and make more deliberate corrections and enhancements. A human is slow and methodical but unconstrained. Automation is fast but significantly more limited in what it can do.”
What we are left with is an affordable alternative to make better-sounding music. Collin Mcloughlin from eMastered commented to The Verge “Rather than replacing jobs or disrupting an industry we can see ourselves as creating a new market, allowing people who currently can’t receive quality mastering to finally have an opportunity to do so.” He further went on to say, “For the absolute best mastering, however, a traditional mastering engineer will always be the best ultimate option.”
It is difficult to say if Artificial Intelligence will ever employ the same nuance a person does toward music, though it may not necessarily need to. Artificial Intelligence models for mastering have been developed enough to be an affordable option for the majority of musicians. Pillion says, “To the people who don’t believe that AI can make competitive sound, I’d say the proof is in the ten million tracks we’ve mastered for millions of artists around the world.” He further says .”I’m sure that when the automatic camera was introduced, people had their doubts, but no one can argue that it hasn’t earned its place in the creative field.”
Read more about the applications of Artificial Intelligence here
Real-Life Applications of AI....
17 min read
Real-Life Applications of AI....
24 min read