By SOUNDSLANKA Team — November 7, 2026
You do not need to play an instrument. You do not need to read music. You do not need a studio, a microphone, or a single dollar. What you need is an idea, a free account, and about an hour. By the end of this guide you will have a finished song — verses, chorus, vocals, the lot — and you will understand every step well enough to do it again tomorrow without me.
The direct answer: Write or generate lyrics, paste them into Suno's Custom Mode (or describe the song in Simple Mode), generate two versions, pick the best, extend it to full length, split it into stems, and finish it in a free DAW like BandLab — all free on Suno's 50-credits-a-day plan.
What Do You Need Before You Start?
Almost nothing, and that is the point. A computer or a phone with a browser, a free Suno account, and headphones. That is the entire shopping list for your first song. If you want to go further — editing, mixing, releasing — add a free DAW (BandLab runs in the browser, Cakewalk is free on Windows, GarageBand is free on Mac) and a free lyrics helper like ChatGPT. Total cost: zero.
One decision to make before you touch anything: what is the song about? Not the genre — the story. "A song about missing the last train home" will always beat "a pop song" because the AI has something to hold onto. Write one sentence about the feeling, the person, or the moment. Everything else in this guide hangs off that sentence.
Step 1: Get Your Lyrics (10 Minutes)
You have three options, and all three are fine.
Write them yourself. Even four rough lines are enough — a verse and a chorus idea. AI tools work best when they have your words to shape, and the song will feel more like yours. Do not worry about perfect rhymes; worry about one honest image. "The platform was empty and the rain had my jacket" beats "baby, you're my shining star" every time.
Ask a chatbot. Tell ChatGPT, Claude or Gemini the story, the genre, and the structure you want — verse, chorus, verse, chorus, bridge — and ask for a draft. Then do the most important part: rewrite at least a third of it in your own words. AI lyrics have tells — "neon lights," "dancing through the night," "whispers in the dark" — and your job is to cut the ones that sound like everyone else's song. We compared the best lyric tools in detail in our guide to AI tools for writing song lyrics.
Let Suno write them. Suno's Custom Mode has an Auto Lyric option that writes lyrics from your prompt. It is fast and often decent, but it is also the least original option — the same prompt gives thousands of people similar lines. Use it to get unstuck, not to finish the job.
Whatever you choose, format the lyrics with section tags in square brackets: [Verse 1], [Chorus], [Verse 2], [Bridge], [Outro]. Suno reads these tags and structures the song around them.
Step 2: Generate the Song in Suno (5 Minutes)
Go to Suno and open Create. You will see two modes, and picking the right one matters.
Simple Mode is for speed: type a description like "a melancholic indie folk song about missing the last train home, gentle acoustic guitar, soft male vocals" and Suno writes the lyrics and the music. Use this for your very first try, just to hear what is possible.
Custom Mode is where real songs happen. Paste your own lyrics, add a style description, and pick your model. Free accounts run v6-mini, which is fast and good. Paid plans unlock v6 and v6-wild for more polish. Set the song title, and if you know it, the style — "indie folk, 90 BPM, acoustic guitar, warm" is a better prompt than "folk."
Hit Create. Suno returns two versions of your song. Listen to both, all the way through. The first ten seconds lie — a great intro can hide a weak chorus, and a slow start can hide a beautiful middle. Judge the chorus above everything else, because that is the part people remember.
Step 3: Pick the Best Take and Generate More (10 Minutes)
Here is where beginners waste their free credits: they fall in love with the first thing they hear. Do not. Generate at least three or four pairs before you commit. Suno's free plan gives you 50 credits a day — about ten songs — so a single session has room for real comparison.
Listen for four things. Does the vocal sit naturally on the words, or does it mangle a line? Does the chorus lift, or does it sound like the verse with more instruments? Does the ending feel finished, or does it just stop? And does the song sound like the genre you asked for, or did it drift into generic pop?
Keep the best version. Suno lets you like and save tracks to your library, and you can rename the keeper so you can find it later. This small habit matters more than it sounds — after a week of generating, your library becomes a mess without it.
Step 4: Extend, Edit and Fix the Weak Parts (15 Minutes)
Almost no AI song is perfect on the first generation. The good news is that Suno's v6 model lets you fix songs in plain language instead of starting over.
The song is too short. Use Extend to add another section — a second verse, a bridge, an outro. Each extension adds up to a couple of minutes, and you can extend more than once to reach a full three- or four-minute song.
One section is weak. Select the section and describe the change in words: "make the chorus bigger, add drums and a vocal harmony." Suno regenerates just that part and keeps the rest.
One lyric is wrong. You can edit a single line without touching the music. Fix the misheard word, the awkward rhyme, the line that makes you cringe.
The whole thing needs a different feel. Use Covers to reinterpret your song in a new style — the same lyrics and melody, reimagined as, say, an acoustic ballad instead of an electronic track. It is the fastest way to learn what your song actually wants to be.
This is the step that separates people who "tried AI music once" from people who finish songs. Budget fifteen minutes here. It is where the song stops sounding generated and starts sounding yours.
Step 5: Split Into Stems and Finish in a DAW (15 Minutes)
A finished-sounding song and a finished song are different things. The final step is taking the track apart and putting it back together properly.
On Suno's Pro plan, Get Stems splits your song into separate vocal and instrument tracks — up to 12 stems with Auto Split — so you can rebalance them in a DAW. On the free plan you cannot split stems inside Suno, but you can still download the finished mix if your account has trial downloads, or use a free stem separator like Ultimate Vocal Remover or the tools inside BandLab.
Import the parts into your DAW and do three simple things. First, balance the levels: the vocal should sit clearly above the music, not fight it. Second, trim the edges: cut the half-second of silence at the start, fade the last two seconds instead of ending on a hard stop. Third, add a touch of mastering: a gentle limiter and a small EQ lift in the high frequencies will make the track sound louder and clearer on phones and laptops, which is where most people will hear it. BandLab's free online mastering does this in one click if you do not want to learn compression today.
Step 6: Export, Name It, and Share It (5 Minutes)
Export as WAV for your archive and MP3 for sharing. Name the file properly — "artist-title-final.mp3" — not "suno-track-482.wav". Write one sentence about the song for wherever you post it: what it is about, and that it was made with AI. Honesty here costs nothing and buys you goodwill; people are far more forgiving of AI's rough edges when you say so upfront.
Where you can share it depends on rights. On Suno's free plan your tracks are non-commercial and Suno retains ownership, so keep them to personal sharing — friends, social posts for fun, your own portfolio. If you want to release the song on Spotify, put it in a monetised video, or sell it, you need a Pro plan download with commercial rights. That is $10 a month, or $8 a month billed yearly, and it is the only honest way to release Suno tracks. If you want a downloadable free-tier file, remake the final version in Mureka, whose free plan includes MP3 downloads — we compare all the free options in our guide to the best free AI music generators.
What Does a Full AI Song Actually Cost?
Here is the honest price list, because "free" deserves specifics.
| Stage |
Free option |
Paid option |
| Lyrics |
ChatGPT free tier, or write your own |
LyricStudio Pro from ~$4/month |
| Song generation |
Suno free: 50 credits/day (~10 songs) |
Suno Pro $10/mo ($8 yearly) |
| Editing and extending |
Included on free plan |
Same, plus v6 and v6-wild models |
| Stem separation |
Free tools (Ultimate Vocal Remover, BandLab) |
Suno Pro: 2 stem types; Premier: 3 |
| DAW and mastering |
BandLab, Cakewalk, GarageBand — all free |
Optional |
| Releasing commercially |
Not possible on free tiers |
Suno Pro $10/mo with commercial rights |
A complete, releasable song costs $10 a month if you use Suno Pro, and $0 if you keep it personal. That is the whole budget conversation.
Common Beginner Mistakes (and the Fix)
Prompting with a genre only. "Make a pop song" gives you the average of every pop song. Add a story, a mood, an instrument, a tempo: "a hopeful pop song about starting over, piano-led, 100 BPM, female vocals." Specific prompts get specific songs.
Accepting the first generation. The first take is a draft. Generate four or five, compare choruses, and keep the best. Your free credits reset daily — use them.
Ignoring the lyrics. The music model will sing whatever you paste, including typos and lines that do not scan. Read your lyrics out loud before you generate. If a line is hard to say, it will be hard to sing.
Skipping the ending. AI songs often end awkwardly. Extend an outro, or fade the last few seconds in your DAW. A clean ending is the difference between a demo and a song.
Forgetting the rights. Do not release a free-tier song commercially. The $10 Pro plan is not a suggestion — it is how the licensing works.
Frequently Asked Questions
How long does it take to make a full song with AI?
About an hour for your first one: 10 minutes on lyrics, 5 on generation, 10 choosing a take, 15 editing, 15 finishing in a DAW, and 5 exporting. After a few songs you will do it in 30 minutes, because you will stop second-guessing every step.
Do I need any musical experience to make a song with AI?
No. You need taste, not training — the ability to hear which of two versions is better. Everything technical, from chords to vocal performance, is handled by the model. Experience helps you prompt better and finish faster, but it is not a requirement to start.
Can I make a full song with AI for free?
Yes, for personal use. Suno's free plan gives you 50 credits a day, enough for about 10 songs, and free DAWs like BandLab handle the finishing. You cannot release it commercially or download it freely on a new free account, so for a releasable song you need Suno Pro at $10 a month.
How long can an AI-generated song be?
Suno's v6 models generate up to 8 minutes in a single pass, which covers almost any song. Most songs end up 2 to 4 minutes after editing. Udio generates about 2 minutes and 10 seconds per pass and needs extending for full-length songs.
Can I use my own voice in an AI song?
Yes, on Suno's Pro plan, which lets you clone your own verified voice and use it across your songs. The free plan does not include voice cloning. Always clone only your own voice, or a voice you have clear permission to use.
Sources
Disclaimer
Suno's plans, credit allowances, download caps and commercial-use terms change frequently — the download cap arrived in September 2026 and prices moved again in November. Check Suno's own pricing page before you pay or release anything. Free-tier songs are for personal use only; commercial release requires a paid plan with commercial rights.
More from SOUNDSLANKA
About SOUNDSLANKA: SOUNDSLANKA is your home for music production — free and premium tools, VST and AAX plugins, loops and samples, KONTAKT instruments, and the latest AI music news. Free tools, honest reviews, and premium picks clearly marked.
Comments
Post a Comment