Soundraw makes instrumentals: what do you use when the song needs words?

You picked a genre blend, pushed the energy up in the second half, trimmed the track to two minutes ten, and exported it. The beat is right. Then somebody asks where the chorus is, and you realise the whole track is waiting for a voice that Soundraw was never going to supply.

That is not a flaw you can prompt your way around. At the time of writing, Soundraw is built for instrumentals: its own page talks about genres, energy, length, a mixer that toggles instruments, and stems, and its homepage features a customer who recorded their own vocals on top of a track. So the answer to the title is two steps, not one tool. First you need the words. Then you need something that sings them, and that is either a generator built to sing your lyric or a person singing over the beat you already have.

The words no generator knows for you

Every tool that sings needs a lyric, and a lyric is the one input that contains facts nobody else has: a name, a street, the thing your sister always says at closing time. Instrumental tools skip that problem because they never have to solve it. The moment you want words, the problem is yours.

This is where we come in, and only here. The lyrics generator on sunomarket takes a short form brief and returns a title and a complete lyric with [Verse], [Chorus] and [Bridge] tags already placed. One generation costs 20 credits, and at 100 credits to the dollar that is $0.20 a lyric. Credits come with a plan, not on their own: the cheapest way in is $5 a month, which brings 500 credits every month (25 lyrics), and lifetime is $19 once, with 1 500 credits (75 lyrics) that do not refill.

What it does not do matters just as much. It does not sing. It makes no audio at all, on any plan. It does not distribute, and it has no connection to Soundraw, Suno or any other company named on this page. You take the lyric to a tool that sings, and that tool is a separate company with its own terms.

Twenty cents is what one lyric costs inside a plan, so a rerun is cheap and you can afford to be picky. The $5 month is the cheapest way in: enough to run a brief for the track you already have, read what comes back, and rewrite the theme until it fits. The $19 lifetime plan makes sense once you know you will keep coming back: the catalogue and prompt generator stay yours, and 75 lyrics come with it, once.

If you are still choosing between music tools rather than trying to finish this one track, the wider map lives in which AI music generator fits your job, and the instrumental specialists sitting next to Soundraw are compared in MusicGPT, Boomy, Soundful and the second tier. This page is narrower. It is for the person who already likes the Soundraw sound and now needs a song.

What Soundraw is good at, so you do not throw it away

People leave a tool too early when it fails at a job it never claimed. Soundraw's controls, as its site describes them at the time of writing, are about shaping an instrumental: blending genres, setting length, changing intensity across the track, muting instrument layers, downloading stems. That is a serious toolkit for a beat under a video, a podcast bed or a demo you intend to sing over.

So the useful question is not "what replaces Soundraw". It is "which half of the song does Soundraw keep". There are two honest answers, and they lead to two different routes.

• Soundraw keeps nothing. You liked its mood, but the song is the deliverable, so you rebuild the whole thing in a generator that sings. Soundraw becomes a reference, not a source.
• Soundraw keeps the beat. The instrumental is already approved, already cut to a video, already paid for. You add a vocal on top of it.

Pick before you write a single line, because the two routes want the lyric shaped differently.

Route one: rebuild the song in a generator that sings

This is the right route for most people, and I would default to it. A gift song, a release, a birthday track: in each case the finished song is the product, and a generator that writes the arrangement around the vocal from the start will fit the words better than any vocal glued onto a beat that was made without them.

What to check in the tool you pick, whichever it is:

1. A field for your own lyrics. Not every generator lets you paste your own words; some only accept a description and write something themselves. You want the version where your text goes in and gets sung. In Suno this is the Lyrics field, and everything in it gets sung except text inside square brackets, which is why the section tags work. The full handoff is in turning a finished lyric into a song.
2. A way to describe sound in words. This is where your Soundraw session is still useful. Every choice you made with sliders can be written down as a description.
3. Commercial terms you can live with. Check the plan page of the tool itself, today, before you publish anything. Terms in this category change often, and free tiers commonly restrict what you may do with the output.

Here is what translating a Soundraw session into a written style description looks like. The left side is what you chose on screen. The right side is text you would type into a style field.

What you set in Soundraw           What you write instead
-------------------------------    ---------------------------------------
genre blend: lo-fi + jazz          lo-fi hip hop, jazzy Rhodes chords
energy: low, up in second half     laid-back, lifts in the second half
instruments: drums down, bass up   sparse brushed drums, upright bass
length: about two minutes          short song, verse chorus verse chorus
(no vocal option)                  warm female vocal, close and intimate

Two things about that right-hand column. Tempo is written as a feel (laid-back), not a number, because in Suno a numeric BPM in the style field tends to be ignored or to fight the other words, and other tools may treat it differently. And the vocal line is the new decision Soundraw never asked you to make. Decide it deliberately: whose voice, how close, how much air.

Our prompt generator builds this kind of style text, but be clear about its scope: it was made for Suno's format. It is not a Soundraw tool and we have not checked it against every other generator. If you use a different singing tool, treat its output as a draft description and adapt it to what that tool's own help pages ask for.

Route two: keep the beat, add a voice on top

Sometimes the beat is locked. A client approved it, the edit is cut to it, the licence is paid. Then you keep it and put a vocal over it, and you have two ways to get that vocal.

A person sings it. You, a friend, a hired singer. This is the simplest route to reason about, because no one's audio goes into a second AI service, and it is the workflow Soundraw's own homepage shows. The lyric has one extra job here: it has to fit sections that already exist. A generator that sings will stretch the arrangement to fit your verse. A fixed beat will not.

So before you write, map the track. Play it, count the sections, and note where the energy changes. You set those changes yourself in Soundraw, so you already know roughly where they are.

Section map of the exported beat (counted by ear)

0:00  intro        8 bars    no vocal, let it breathe
0:16  verse 1     16 bars    about 8 short lines
0:48  chorus       8 bars    4 lines, title in line 1 and line 4
1:04  verse 2     16 bars    about 8 short lines
1:36  chorus       8 bars    same 4 lines
1:52  outro        8 bars    one repeated line, or nothing

The line counts are a rough guide, not a law. A short line at a relaxed tempo tends to fill about two bars, and a busy rap line can fill one. What the map gives you is a hard ceiling: a 16-bar verse will not hold twenty lines, and if you write twenty, the singer has to cram or cut on the day.

When you fill the form, this is where the structure field earns its keep. classic gives verses, choruses and a bridge. no-bridge fits the map above, where there is simply no room for a third section. loop suits a beat built on one repeating phrase, where the hook comes back again and again. Then edit the returned lyric down to your line counts. A lyric generator does not hear your beat, so matching bars is your job, and it takes ten minutes with the map in front of you.

A generator adds the vocal. At the time of writing, Suno's help centre describes an Add Vocals feature, in beta and limited to its paid Pro and Premier plans, that layers a vocal from lyrics you provide onto an instrumental you upload; some other tools advertise something similar. Check Suno's current help page before you plan around it. Two warnings before you try it with a Soundraw export. First, uploads get rejected for reasons that are not obvious, which is covered in why Suno rejects an upload. Second, and more important: whether your Soundraw licence lets you feed its track into another AI service, and then release the result, is a question for Soundraw's current licence terms and the other tool's terms, read together. Do not assume that a licence to use a beat in a video covers turning it into a new song somewhere else.

This is not legal advice. Licensing and ownership rules differ by country and by each company's terms, and who owns AI-generated music explains why the question is less settled than it looks. If money or a release depends on the answer, ask someone qualified where you live.

A brief that works for either route

The form has closed fields, and each one changes the lyric in a way you can predict. Here is a filled brief for the lo-fi café track from the example above.

POV        first
language   en
structure  no-bridge
hook       title-line
vocal      female
emotion    calm
age        25-34, 35-44
theme      Opening video for my sister's café, named Linden Row after
           our grandmother's street. Early morning, chairs coming down
           off the tables, cardamom buns in the oven, the first regular
           who always orders two.

That theme is 202 characters, under the 300 limit, and every fact in it is something the generator can actually use: a name to build the title around, a place, a smell, a person. Whatever comes back can only draw on what is in that box. If you want the grandmother in the chorus, she has to be in the theme; the generator cannot guess her.

A chorus built from that brief might open like this, and you can see where each line came from:

[Chorus]
Linden Row, the chairs come down
Cardamom is in the oven
Two for the one who's always first
Linden Row, my grandmother's street

Every noun in it comes from the theme: the name, the chairs, the cardamom, the oven, the street. That is the test worth running on anything you get back: if a line mentions something you never wrote down, it is filler, and you should either cut it or add the real detail to the theme and run the brief again for another 20 credits.

The hook field decides where the weight lands. title-line builds the chorus around the title line, so the name keeps coming back, which is what you want when the name is the point, as it is for a café. chant gives a short phrase built to repeat, good for a beat-driven track where the chorus is more rhythm than story. question and confession suit songs addressed to a person. There is no duet option in vocal: it is female, male or unspecified, so if you want two voices, write two parts yourself and mark them in the lyric.

The four mistakes that cost the most time

Writing before choosing the route. A lyric written for a generator that sings will usually run long for a fixed beat. Choose route one or route two first.

Pasting Soundraw's settings into the lyric. Words like energy: high or lo-fi jazz belong in a style description. In Suno's Lyrics field, anything outside square brackets gets sung, so a stray setting becomes a sung line.

Expecting the lyric generator to hear the beat. It reads your brief and nothing else. It cannot count your bars, so the section map is on you.

Assuming one licence covers two companies. A Soundraw beat plus another tool's vocal is two sets of terms. Read both before you publish.

FAQ

Does Soundraw make songs with vocals? At the time of writing its site presents it as an instrumental generator with controls for genre, length, intensity and instruments. Features change, so check Soundraw's own current page before you rely on this either way.

What should I use instead of Soundraw when the song needs words? A generator that sings your own lyric, for most people. Write the lyric first, then describe the sound you liked in Soundraw as words in that tool's style field.

Can I keep my Soundraw beat and add vocals? With a human singer, that is the simplest path, and Soundraw's own homepage carries a testimonial from a customer who did exactly that. Still check that your plan's licence covers the release you have in mind. With an AI tool that adds vocals to uploaded audio, it depends on both companies' current terms.

Does sunomarket make the audio? No. It returns a title and a lyric with section tags. The audio happens in a separate tool you choose.

Is sunomarket connected to Soundraw or Suno? No. Both are separate companies. We do not resell, represent or control either.

What does a lyric cost? 20 credits, which is $0.20 at 100 credits to the dollar. Credits come with a plan: $5 a month brings 500 credits (25 lyrics) every month, and $19 once for lifetime brings 1 500 credits (75 lyrics) that do not refill. There is no free trial and no way to pay for a single lyric.

Do this now

Open the track you exported and write the section map: where the intro ends, how many bars each verse runs, where the chorus hits. Then open the lyrics generator, pick no-bridge if your map has no room for a third section, put one real name and two real details in the theme, and run the brief. The $5 monthly plan is the cheapest way in and covers 25 runs like this one. If the chorus that comes back has your name in it and fits your eight bars, and you know you will do this again, the $19 lifetime plan keeps the catalogue and prompt generator for good and comes with 75 lyrics.