Dashboard & Core Workspace

Overview

Stable Audio is powered by Stability AI's Stable Audio 3 text-to-audio model and is a production interface designed to streamline music and sound design creation.

Unlike traditional Digital Audio Workstations (DAWs) that require track routing, virtual instruments, and complex plugin chains, Stable Audio generates full mixes or individual stems directly from natural language prompts and structural toggles.

Workspace Breakdown

1. Left Sidebar (Navigation & History)

Recent Generations: Lists your recently generated audio jobs along with their relative timestamps.
Credits Counter (Bottom Left): Tracks your remaining generation credits in real time.

2. Prompting Engine (Center Workspace)

The central input area translates written descriptions into audio compositions.

Text Input Box: Enter descriptive text specifying genre, tempo (BPM), instruments, mood, and production characteristics.
- Example prompt: sultry tropical house, chill afro-dancehall, smooth pop R&B, mid-tempo 105 bpm, heavy muted bassline, warm analog synths, crisp percussion, gentle acoustic guitar riffs, plucky melody, nocturnal intimate vibe, late-night dance rhythm
Sample Presets (Inspiration Cards): Quick-start prompt templates located beneath the generation bar.
- Click any preset card to populate the prompt field.
- Click ↻ shuffle samples to load a new set of sample cards.

3. Generation & Output Controls

Located directly below the main text prompt, these controls define how your audio is rendered:

Control Feature	Selection Options	Description & Use Case
Output Type	`FULL MIX` `STEMS`	`FULL MIX` creates a single combined audio file. `STEMS` renders isolated audio components (e.g., drums, bass, synth) for further mixing or export into another DAW.
Length Selector	`0:30 LOOP` `1:00 SHORT` `1:30 SONG` `3:00 EXT` `AUTO`	Sets the output duration. Use `0:30 LOOP` for seamless repeating beats or ambient beds, and `EXT` / `SONG` for full-length structural arrangements.
Direct Full Mix	Dynamic Label	Displays the active render path based on your selection. When `FULL MIX` is selected, it displays `DIRECT FULL MIX` (indicating direct single-file output). When `STEMS` is selected, it updates to display the active stem/diffusion parameters (such as step configuration or model engine).
Action Button	`START →`	Triggers the audio generation job. Shortcut: `⌘ + Enter` (macOS) or `Ctrl + Enter` (Windows/Linux).

Quick-Start Workflow

Type a Prompt or select a Sample Card.
Set your Output Type (FULL MIX or STEMS).
Choose your desired track duration from the Length Dropdown.
Click START → or press ⌘ + Enter to process your request.