How to generate background music and ambience

What this tool makes

Sound & Music composes instrumental music beds and atmospheres from a plain-language description: a lo-fi study bed, rain on a tin roof, a slow cinematic rise. The composer is Lyria 3, a music model currently in preview, and the studio carries a Preview pill because of it: output and availability can shift while the model matures.

It is not a voice tool, and it is not a one-shot sound-effect tool. Single effects (a door slam, a footstep) are coming, but honestly are not here yet.

Writing a good prompt

Prompts run from 5 to 600 characters, and short ones work well. The three ingredients that steer the model most are mood, instruments, and pace. Name all three and you will usually land close on the first try.

• Mood: calm, tense, warm, triumphant, melancholy.
• Instruments: mellow electric piano, strings and deep percussion, a soft ambient pad.
• Pace and shape: relaxed tempo, slow build rising to a bright peak, steady and unhurried.
• Real starter prompts from the studio: "Warm lo-fi hip hop beat with mellow electric piano and vinyl crackle, relaxed tempo" and "Gentle rain on a tin roof with distant soft thunder, calm and steady".

Clip length and generation time

The tool produces short clips; that is the only length offered right now, and longer-form pieces are not available yet. Composing takes about 20 to 60 seconds, shown as an honest status line rather than a fake percentage bar.

Output, saving, and rights

Each generation is a finished MP3, played back in the studio and auto-saved to your library, tagged with the engine name and your prompt, so you can find it again later. It is yours to download and use in your voice work, the same ownership terms as everything else you make here.

Generating music requires sign-in, and a generation is recorded against your monthly allowance by the prompt's character count.