Suno: songs from a text prompt
Generate a full song with vocals from a description, and understand what the licence does and does not cover.
Pricing
Free tier with credits; paid plans from about US$10/month (check the official page)
Official site
Open the official website (opens in a new tab)
Prices and plans change. Check the official page before you pay.
Suno generates music, including vocals, from a text description. A prompt like "acoustic folk song about missing the last train, warm male vocal, 90 bpm" produces a complete track in under a minute.
What it is good at
- Complete drafts. Structure, instrumentation, vocals and mixing in one step, which is remarkable for a demo or a podcast intro.
- Style range. From lo-fi to metal to film score, with the genre conventions recognisable.
- Lyric setting. Paste your own lyrics and it sings them, which makes it useful for songwriters testing a melody.
- Iteration speed. Ten variations in ten minutes, which is faster than any other way to explore a direction.
Where it fails
- Lyrics drift. Words get repeated, mispronounced or invented, especially in fast passages.
- Structure. Verses and choruses can blur; a request for a bridge often produces another chorus.
- Mix quality. Tracks sound loud and compressed, and stems are not always clean enough for a real production.
- Originality. Output follows the patterns of its training data closely; a generic pop prompt produces generic pop.
Using it well
- Describe genre, mood, tempo and instrumentation, then add one distinctive detail.
- Write your own lyrics for anything that will be listened to twice.
- Use it for a demo, then re-record with real musicians if the song matters.
- Keep the prompt with the file; regenerating the same idea later is easier with the notes.
Licensing
Free-tier output is for non-commercial use and the terms have changed more than once. Paid plans grant commercial rights for the tracks you generate, with conditions. Read the current licence before putting a track in an ad or a paid course, and keep a record of the plan you were on when you made it.
Who it is for
Podcasters needing an intro, game developers prototyping a soundtrack, songwriters exploring ideas, and teachers making a memorable recap song. It is not a replacement for a composer when the music is the product.
Alternatives
Udio is the closest competitor with a similar feature set and different strengths in vocals. For royalty-free library music, Epidemic Sound and Artlist avoid the licensing questions entirely. For sound effects and stems, dedicated tools do a better job than a song generator.