Annual Membership 50% OFF - Limited Time!
Upload a beat, a backing track or a finished instrumental and describe the singing: paste lyrics or a vocal brief, choose a style and the AI performs a vocal that follows the music.
Preset: style pop, warm expressive lead vocal, with negative tags off-key, harsh distortion, spoken word. Add a title, your lyrics or brief, and upload the instrumental.
Many producers have folders of instrumentals waiting for a singer. Hiring a vocalist takes time and money, and a rough scratch vocal is often enough to decide whether the beat is worth finishing. This page makes the singer: upload the music, give it words, and it generates a vocal performance that fits the harmony and tempo of the track.
The output is a complete song with the vocal included, so you can judge melody, flow and mood immediately.
Two inputs matter most: the lyrics and the style. For lyrics, short lines with a clear rhyme scheme are sung more reliably than dense prose. Mark sections with headings such as verse and chorus so the vocal repeats the hook.
For style, say what kind of singer you want: breathy, belting, soulful, rap-sung. The exclusions field keeps unwanted characteristics out; the pre-filled list steers away from off-key and distorted delivery.
Beyond the basics there are sliders that balance how closely the model follows your style description, how creative it is allowed to be and how strongly it responds to the audio you uploaded. If the vocal ignores the groove of your beat, raise the audio weight; if it sounds too safe, raise the creativity.
Keep the upload clean. Muddy or clipping instrumentals give the model a poorer picture of tempo and key, and the vocal will wander. A bounce with a little headroom works best.
The generated vocal follows the style and lyrics you give it, but it can mispronounce unusual words or rush dense lines. Gender is a preference rather than a rule, and the same lyrics produce different melodies on each run. The title, lyrics, style and negative tags fields are all required, so fill in each one, even with a short phrase. Uploads should be music without a lead vocal.
Limits and cost for this page, read from the model configuration.
| Model | Suno add-vocals (V6 Mini by default) |
|---|---|
| Input | One instrumental or music audio upload |
| Required fields | Title (100), lyrics or guidance (5,000), style (1,000), negative tags (500) |
| Controls | Vocal gender (auto, male, female), style weight, creativity, audio weight, model |
| Cost | 9 credits per generation |
| Result | A full song with the generated vocal over your music |
Jobs for a generated singer.
Hear whether an instrumental works with vocals before booking a singer.
Check your lyrics and melody feel against a finished backing track.
Add a sung hook to a cue you made for a project.
Run the same lyrics in different vocal styles to find the right voice.
Upload, write the brief, generate.
Add one music or instrumental audio file with no lead vocal on it.
Use the karaoke maker to strip an existing vocal first
Fill in the title, the lyrics or vocal guidance, a style, and anything to exclude. The form opens with a pop style and exclusions to edit.
Put a short chorus first if you are pasting full lyrics
One run costs 9 credits. Set a preferred vocal gender if you like, then listen and try again with a different brief.
Lyrics with simple rhymes sing more reliably
What the form needs and what you get.
Pages for the other half of the song.
Choose the plan that's right for you. No hidden fees, no surprises.
Start your music journey
Unleash your creative potential
Professional-grade music production
