Mixing and Mastering as Communication: What Are You Really Saying With Your Sound?

People around sometimes tell me that AI advances will replace professionals and end my industry. This is a widespread concern across all creative industries today. Every time somebody tells me that AI will take my job, I have to describe to them what I do from the beginning.

Really, what is mixing? What are the behind-the-scenes activities of audio engineers that contribute to the greatness of the songs we love? If we have AI that can put everything together perfectly or even replicate the biggest mixing engineers in the world, wouldn’t it be efficient?

AI in Music isn’t New

To answer those questions, I believe we have to go back a couple of years, and we have to remind ourselves that creative AI was already making endless songs in the style of famous composers in just 2017. It’s 8 years ago! To have a reference, the latest iPhone was the iPhone 8 that year. It’s the year Kendrick Lamar released the DAMN. album. I’m telling you this because artificial intelligence is not a new thing in music.

But what is the relationship between the subject of this article and the advancements of artificial intelligence? Well, if you believe that the job of mixing engineers is making everything wonderful and sound awesome, you’re absolutely wrong. It wouldn't be considered an art form if it were that simple. People wouldn’t dedicate years to strengthening their skills.

Making a good-sounding record in 2025 is easier than ever. Samples and loops sound amazing on their own. Synth presets sound amazing. Guitar amp simulations and their presets sound amazing. The basic sounds of almost any DAW now sound so incredibly good that you can definitely create a song at the same level as a Grammy Award-winning song with a couple of hundred dollars of software investment.

AI Music Tools and Their Limits

Another thing that is coming up now is the AI tools that make a song from scratch in just a matter of seconds. The sound quality is bad now, but I think they will improve and be able to make good records alone.

But think about the songs we love the most. Think about the songs we repeatedly listen to. Are these the ones that sound the greatest? I’ll tell you no. The ones that we have great memories and connections with are the ones that can communicate correctly with us. This is why communication is the most important aspect of mixing and mastering, and engineers spend years refining their skills.

I love how Tchad Blake, the mixer of Arctic Monkeys’ excellent album “AM,” approaches mixing. The multitracks sound great when they arrive at his desk, and he deliberately plays with them until he spoils them enough that it starts to sound like something special of its own. He doesn’t try to make them sound better. He doesn’t even think about sounding great on his own. He works on making an emotional connection with the song itself while adding ugliness to it, and he stops when he can be pleased with the communication that he created with his own technical skills.

Mixing is Not Just Technical Skills

That’s another aspect of making great-sounding recordings. Somehow, every beginner and every outsider thinks that mixing is a technical skill; therefore, they can learn how to mix better by just learning the technical tools, such as EQ, compressor, reverb, delay, or more fancy stuff like M/S processing or parallel compression.

But the technical skills are only one aspect of mixing and mastering. Someone who knows compressors can ruin a song with them, but someone with good ears can improve it without knowing how they work. ‘Great ears’ are not the ones that can hear every resonance frequency right away or can pick up every nuance immediately, but they are the ears that can decide what is best for that particular song. This is why certain producers, such as Rick Rubin, get paid even though they lack technical skills.

Yes, a compressor can make a song incredibly punchy, but does that song really need to be punchy in the first place?

If there were only one great-sounding drum, one great-sounding compressor setting, or one optimal frequency spectrum for a great-sounding song, that would be a very, very boring job for what we do as mixing engineers. But it’s a very fun job that I have to think about the consequences of every decision. Sometimes the song needs very punchy drums for sure, but sometimes they need to be buried in the mix, and sometimes they have to be muted in some sections to make the next section even bigger.

Mixing Decisions are Communication Decisions

Those are all puzzles that we have to solve as mixing and mastering engineers and producers. The best decision for a song will always be the one that can communicate with the audience better.

If it’s a song that you want your audience to sing along to, turn up the volume on the vocals a bit more. If it’s a song that you want your audience to really feel the drums and the low end, do the opposite so they can increase the volume to hear the vocals. If you want your audience to feel the vocals more intimate, reduce all the reverbs and delays or mute them completely in the mix. If you want the audience to feel the vocals as divine as possible, increase all of them oppositely.

I always used this technique at live concerts. Some artists want to be close to their audience and make jokes in between songs. So what I do on these occasions is to reduce all the effects immediately after a song finishes. But occasionally it’s the opposite. The artist, as an introverted rock star, prefers not to speak on stage and only says "thank you" between songs or shares a few sentences at most. On that occasion, I don’t touch any effects to maintain the distance between the artist and their audience.

Or think about it this way. Think about a song that I want to communicate with the audience as a lo-fi, bedroom record. Intimate, with the right amateur soul, good intentions, and a low budget. If I send this song to a mastering engineer and the first thing they do is try to fix the issues in that recording and increase the high frequencies to the mainstream degree, they will ruin the whole idea. They should approach it as it is and continue the narrative that has been done throughout the whole production, and respect the communication decisions.

Mixing: Might be Simpler but not Easier

Those are all decisions we have to make while mixing a record. It definitely doesn’t make the whole process easier, but it makes it simpler. You can connect the audience to the song better by focusing on the more important aspects instead of the technical ones that no one cares about.

This does not imply that the technical aspects are of lesser importance. Knowing how to use a compressor intuitively eliminates the need for extra thought and allows you to act on your instincts immediately. You don’t spend hours finding the best settings. Mixing is not a simple technical process; it involves more than just using fancy-looking faders and having fun. It is both.

Coming back to the artificial intelligence topic, I don’t think they can take my job because they will not know how to create unique sounds for a song that can communicate with the audience better. It can create a great-sounding record, but when everybody can easily make a great-sounding record, there won’t be any point in making them anymore. And people will always need us to trust them with their sound to create real connections with their audiences.