Here is a detailed transcription of the video "Make Better Suno Songs with Square Brackets":
Want your Suno songs to sound complete? Square brackets are the key. In this video I’ll show you how to use them to define verses, choruses, bridges, vocal styles, interludes, refrains, drops and more — giving your AI-generated songs a real structure.
Ever wonder why your Suno songs sound unfinished? The secret is in the square brackets. They unlock pro level song structure in seconds. In this video, I'll show you exactly how to use square brackets to map out your song parts — verses, choruses, bridges, and more. Once you learn this trick, you'll be able to guide Suno into creating complete polished tracks that sound like real songs, not just random lyrics stitched together.
Let's start with core song sections. Most modern songs have some variation of this pattern: an intro, a verse, to either a pre-chorus or directly to a chorus. And then it'll cycle back to verse two, pre-chorus, chorus, go to a bridge, and then come back to a third chorus, maybe even a fourth chorus, and then there's an outro. These are the main parts of most modern songs you hear on the radio or Spotify.
But let's say you want to add something additional to this song. You can add things in between these main sections. Once you have your core song sections figured out, maybe you'll want to add a little more to the framework. In that case, you can add instrumentals or arrangement elements — such as a guitar solo, a piano solo, sax solo, drum breaks, there are all different kinds of things you could add in between, for instance, the chorus and verse two, or anywhere along here. It's kind of up to you.
Usually, I'll throw in something after the first and second chorus just to clean the palette. I might add a piano solo, a breakdown, maybe a rise or a ramp. That brings a lot more interest to your song framework and suddenly you have your own unique framework.
Now that we have our main sections figured out — what order we want them to be in — maybe we want to give a little more detail within each section as to how the vocals will sound. That's where we can add vocal-specific sections. Now, typically they aren't independent like this. Typically, we'd go to the chorus and then say: chorus, falsetto, lead, for instance, combining the section header with the type of vocals we want to hear in that section. Or, in verse one, maybe we have spoken word or rap vocals.
We can add to each of these main sections of the framework — we can add specific vocal types. Gang vocals, maybe later in a chorus, maybe instead of just the main singer, you want a group of people singing. Gang vocals. That's an option. For the bridge, maybe you want a whispered vocal or a chant for the bridge. If you're doing other styles of music, like blues, you might want to do call-and-response. For instance: verse one is going to have a call-and-response feel — line one is one person, line two is another person responding to it.
Having the ability to determine the vocal type in each part of your song gives a lot more variety and creativity. Most songs have these main song sections: verse, pre-chorus, chorus, bridge, intro, outro. But some types of music don't use these sections. Some types add an interlude, for instance. In orchestral music, it's often called an intermezzo passage. In jazz there may be solos between verses and choruses. In hip-hop, there might be spoken word. As you move from a rap, it might go back to spoken word, e.g., "I never knew they felt like this about me until I saw..." and that could also be true for R&B.
Sometimes they add skits: little vignettes within the song, like "Oh no, the police are coming, let's run and hide," or "We're gonna get these guys." These are kinds of interludes you might want to add into your core song sections depending on the type of song. Progressive rock sometimes has a calm instrumental before a heavy section. Pop is similar to hip-hop and R&B — they might have a spoken word or instrumental.
Interludes can add more uniqueness to your song. Some songs don't use choruses so much as they use a refrain after a verse — typically one or two lines long, used instead of a longer chorus. It's more like a tagline after each verse. Folk, blues, early pop, rock stories work well with refrains. Hymns, spirituals, and some country music use refrains. You're just swapping out the chorus, so you could use a refrain instead.
Another genre with a different framework is EDM/dance/techno. For an EDM song, you'd probably use: [buildup], [drop], and [breakdown]. Under [buildup], you’d have your lyrics, same for [drop], same for [breakdown], just like with chorus sections.
In summary, square brackets are great because they define the song structure and help specify what type of vocal performance goes with each section. You just combine the song section and vocal type in one square bracket.
Now, let's go over to ChatGPT and put a song together using these elements. Let's specify the core song sections:
-
Intro
-
Verse 1 (rap vocal)
-
Pre-chorus
-
Chorus (falsetto male vocal)
-
Guitar instrumental
Next, copy and paste, change verse one to verse two (rap vocal), pre-chorus (same), falsetto (same), guitar instrumental, and piano solo (solo or instrumental to see both). Then a bridge (spoken word), a third chorus (choose vocal style), outro (whispered vocals).
That’s our framework for ChatGPT. Let's see what it comes up with for lyrics and style. For style, ask ChatGPT for a 300-character style for the song. Copy and paste this into Suno's style field.
In Suno, go to create > custom tab, use 4.5 plus, paste in the lyrics, select style, and the appropriate vocal.
Render a couple songs and see if the square brackets work. Follow along:
-
Intro: "This is an intro, Oh, did the Os hit the pavement, Sun on my back..."
-
Verse (rap vocal): sample lyrics
-
Pre-chorus: sample lyrics
-
Chorus (falsetto male): "I’m flying past the pool of heart shadow, ooh, city never feels too hard with you..."
-
Instrumental: Suno rendered electric guitar instead of specified classical guitar (sometimes does not match instrumental type)
In the second verse and chorus, similar structure is kept. For the bridge, it provided a spoken word section, though sometimes it blends singing and spoken word. For whispered vocals/outro: Suno did not accurately provide a whispered style.
So, square brackets help guide Suno, but don't guarantee you’ll always get exactly what you want — accuracy is about 50 to 75%. Sometimes you have to render multiple times to get the result you’re after.
In conclusion, square brackets give control over Suno’s output. By labeling sections, it encourages natural song structure and energy shifts. It's a small detail but makes a big difference. For more Suno and AI music tutorials, check out my other content.youtube
