Glossary

Product & workflow

Fields, models, and features

Style

The prompt field that steers sound; genre, instruments, vocals, mix. See Lyrics vs Style.

Lyrics / Lyric prompt

Words and structure tags the singer follows; also shapes sections.

Model / version

Generation capability (e.g. v5.x). Newer versions may interpret prompts differently; re-test old lines.

Extend / Continue

Continue a generation from audio; wording in the app varies. Useful when a track cuts before your written outro.

Personas

Suno’s saved voice/vocal identity features (as labeled in-app); pair with clear vocal words in Style.

ReMi

Remix / edit-style flows in the product (name may vary). Treat as another pass over audio, not a substitute for clear Style.

Credits

Currency for generations; check Suno’s official pricing; free tiers often have daily limits.

Production terms (prompt context)

Mix and groove vocabulary

Sidechain

Ducking: one sound dips when another hits (e.g. kick vs bass). Say “sidechain-heavy pads” in Production.

Four-on-the-floor

Kick on every beat; common in house and pop dance. Genre or groove hint.

Double-tracked

Same part recorded twice for width; hint with “doubled vocals” or “wide doubles” in Style.

Suno terminology explained

Key concepts you need to know

Understanding Suno's terminology is essential for effective prompting. The platform uses specific terms for its input fields, generation modes, and output features, and confusing these terms can lead to unpredictable results. This glossary covers the most important terms you will encounter when using Suno and when reading this guide.

Input fields

Style, Lyrics, and Exclude

Style

The text field where you describe the sonic characteristics of the music you want: genre, instruments, mood, vocals, and production. This is the primary control for shaping the sound of your generation. Everything from "acoustic folk, fingerpicked guitar, warm" to "dark techno, 140 BPM, acid bassline, industrial" goes here.

Lyrics

The text field where you enter the words to be sung, along with structure tags ([Verse], [Chorus], etc.). When left empty, Suno either generates an instrumental track or auto-generates lyrics depending on the mode setting.

Exclude

The negative prompt field (under Advanced Options) where you list elements you do not want in the generation. See the Exclude guide for details.

Style Tags

Individual keywords within the Style field, such as "ambient," "808 drums," or "breathy falsetto." Each tag suggests a sonic characteristic. The Reference hub groups larger style-tag lists by category.

Structure Tags

Bracketed labels placed in the Lyrics field to mark song sections: [Verse], [Chorus], [Bridge], [Outro], etc. These tags tell the model how to structure the song. See Structure & tags for the full guide.

Generation modes

Song, Instrumental, and Extend

Song mode

The default generation mode that produces a complete track with vocals (if lyrics are provided) or auto-generated vocals (if the lyrics field is empty). Song mode generates approximately 2-4 minutes of audio depending on the model version.

Instrumental mode

A generation mode that produces music without vocals, regardless of whether lyrics are entered. In instrumental mode, the Style field does all the work: mood and production tags become especially important because there are no vocals to carry emotional content.

Extend

A feature that continues an existing song from a chosen point. Extend can test a new ending, add a section, or make a track longer. It retains part of the source as context, but musical and production continuity can still vary.

Custom mode

The input mode that gives you separate fields for Style and Lyrics, as opposed to the simplified mode where you enter a single prompt description. Custom mode is recommended for all serious prompting because it gives you explicit control over both inputs.

Output and quality terms

Understanding what you get

Generation / Take

A single output from Suno's model. Each generation is unique due to the model's stochastic nature. A single prompt can produce very different results across multiple generations, which is why testing with 3-5 takes is recommended.

Stochastic

The property of being randomly determined. Suno's model is stochastic, meaning the same prompt produces different results each time. This is a feature, not a bug: it allows you to explore variations by re-rolling.

Seed

The internal random value that determines a specific generation. If you could control the seed, you would get identical results from the same prompt. Suno does not currently expose seed control to users.

Extend point

The timestamp from which an Extend operation continues. Choosing the right extend point is important for seamless continuation: typically the last bar of the existing generation.

Prompting concepts

How to think about your inputs

Pillar

One of the five categories of Style tags used in this guide's framework: Genre, Instruments, Mood, Vocals, and Production. Each pillar addresses a different aspect of the sonic recipe. The Prompt Builder is organized around these five pillars.

Style line / Style string

The complete text in the Style field, combining tags from all five pillars into a single comma-separated string. Example: "indie folk, fingerpicked guitar, warm string quartet, bittersweet, close-mic solo vocal, cozy reverb."

Copy-ready snippet

A pre-written Style line designed for a specific sound, available as a one-click copy button throughout this guide. Snippets are starting points: modify them to fit your specific needs.

Contradiction

A conflict between two tags in the Style field, or between the Style field and the Lyrics field. Contradictions (e.g., "whispered" and "belted" in the same Style) are the most common cause of unpredictable Suno output.

Implied element

An instrument, vocal style, or production trait commonly associated with a genre label and therefore possible even when you did not name it. For example, a funk request may bring horn parts. Use Exclude as a focused request and verify whether the element remains.