Suno AI Tutorial Part 1: Make Your First Song (No Music Background)

Suno turns a single sentence describing a song into a finished track, complete with vocals, melody, instruments, and arrangement, generated fresh each time rather than pieced together from existing recordings. Rohan Vijayakumar walks through the first of three training videos for Lyrical Literacy volunteers, aimed at people with no music background and no prior AI tool experience.

What Suno is actually doing

Suno is a generative model trained on a large catalog of music. During training it learned the patterns that link words to sound, what makes a track read as folk rather than hip-hop, hopeful rather than tense. Given a description, it predicts audio that fits that description a segment at a time, each segment continuing from the one before it. That segment-by-segment generation is why the same prompt never returns quite the same song twice.

Signing in with Discord for pro access

Getting started means joining the Humanitarians AI Discord through the invite from a team lead, then going to suno.com and choosing Discord as the sign-in method. Signing in with the same Discord account used to join the server is what links an individual account to the organization's pro plan, and pro access applies automatically once signed in that way.

Finding your way around the workspace

The left sidebar has four items that matter for a first song. Home shows a feed of recent activity. Explore shows tracks other people have published, useful for hearing how a description turns into sound in practice. Library holds every track generated, saved automatically. Create is where songs actually get made. The create page itself has three tabs: simple, advanced, and sounds. Advanced controls lyrics and song structure and is covered in part two of the series. For a first song, the simple tab is where everything happens, and only three things on it matter: the song description box, an instrumental toggle for a track with no vocals, and the create button.

Writing a description with three ingredients

A working description has three parts: style, what genre or sound it should be, such as folk, pop, or hip-hop; mood, what the music should feel like, such as hopeful, playful, or calm; and topic, what the song is actually about. Combined into one sentence, an example given is a gentle acoustic folk song, warm and hopeful, about children discovering the joy of reading. Style, mood, topic, in one sentence, is a complete description ready to submit.

Generating, comparing, and downloading

Clicking create generates two versions at once, usually within 30 to 60 seconds. Each generation costs five credits, and the pro plan carries 2,500 credits a month, roughly 500 songs, which is enough to iterate rather than needing to get the wording right on the first try. The two resulting cards can each be played and compared, since they interpret the same description differently, and listening to both shows what the wording actually produced. The three-dot menu on each card holds remix, edit, share, and download options, and downloading as an MP3 is how the finished file gets saved.

Key takeaways

• Suno generates new audio from a text description rather than combining existing recordings, so the same prompt can produce different results each time.
• Signing into suno.com with the Humanitarians AI Discord account automatically applies the organization's pro plan credits.
• A complete description has three ingredients: style, mood, and topic, combined into one sentence.
• Each generation produces two versions for five credits total against a 2,500-credit monthly pro allowance, which supports iterating rather than aiming for a single perfect result.
• The three-dot menu on a generated track handles remixing, editing, sharing, and downloading as an MP3.

Who this is for

This is for Lyrical Literacy volunteers, or anyone else, with no prior music production or AI tool experience who wants to generate a first complete song in Suno using simple mode.