Create Full-Length Songs with Lyria 3 Pro

Seedance 2.5 IS HERE · Nano Banana 2 & GPT Image 2.5 UNLIMITED UNTIL September 30Upgrade

Overview

Lyria 3 Pro turns a written description into a full song, up to 3 minutes long, without requiring any musical training or production software. Describe the genre, mood, instruments, and structure, and the model builds an audio track around those details. You can also feed it reference images and it will draw on the visual atmosphere when composing. On Picasso IA, the whole process runs directly in your browser, no plugins or downloads required. This makes it a practical option for content creators, game developers, filmmakers, and anyone who needs original music without hiring a composer.

How It Works

Write a text prompt describing the song you want: include genre, mood, instruments, tempo, and any lyric lines you have in mind
Add structural cues like [Verse], [Chorus], and [Bridge] to shape how the song is arranged
Use timestamps such as [0:00 - 0:30] to control which sections appear at specific moments in the track
Optionally upload up to 10 reference images to give the model a visual atmosphere to compose around
Submit the prompt, wait for generation to finish, and download the audio file

Frequently Asked Questions

Do I need programming skills or technical knowledge to use this? No, just open Lyria 3 Pro on Picasso IA, adjust the settings you want, and hit generate.

Is it free to try? Free generations are available so you can test the model before committing to a paid plan. Check the pricing page for current credit details.

How long does it take to get results? Most tracks are ready in under a minute. Generation time can vary slightly based on song length and the complexity of your prompt.

Can I include lyrics in my prompt? Yes. Write the lyric lines directly in your prompt alongside structural tags like [Verse] and [Chorus]. The model will follow the text and arrange vocals around the structure you describe.

What output formats are supported? Lyria 3 Pro returns a downloadable audio file. No additional conversion tools are needed before you can use it.

How many times can I run the model? You can generate as many tracks as your credit balance allows. If you are iterating on a track, adjust one element at a time to close in on the result you want without burning through credits.

Where can I use the outputs? The audio you create is yours to use in videos, games, podcasts, or client projects. Check your account plan for the specific usage terms that apply.

Credit Cost

Each generation consumes 2 credits

2 credits

or 10 credits for 5 generations

Features

Everything this model can do for you

Full-length output

Generate songs up to three minutes from a single text prompt.

Structural control

Use [Verse], [Chorus], and [Bridge] tags to set the exact section layout of your song.

Image-to-music input

Add up to ten reference images to shape the mood and texture of the generated track.

Timestamp precision

Lock specific sounds or transitions to exact moments using time-coded instructions in your prompt.

Genre and instrument control

Specify genre, tempo, lead instruments, and emotional tone to get a track that fits your project.

Lyric integration

Write original lyrics directly in your prompt and have them sung in the final output.

No setup required

Write what you want to hear in plain language and download a finished audio file.

Use Cases

Generate an original soundtrack for a YouTube video by describing the mood, genre, and tempo you need

Turn a landscape photo into an ambient background track for a meditation app or relaxation video

Write a custom jingle for a brand by specifying the product, tone, and lead instruments in a text prompt

Produce a full song with verses, a chorus, and a bridge by laying out the structure directly in your prompt

Create a unique in-game music track for a specific scene by describing the action and atmosphere

Score a short film by uploading reference images and describing each scene's emotional arc

Build a catalog of original tracks in a consistent style for podcast intros, ads, or episode transitions

Examples

56.8s

[0:00 - 0:15] Intro: Soft lo-fi beat with muffled vinyl crackle and distant rain sounds. [0:15 - 0:45] Verse: Warm Fender Rhodes piano melody with gentle bass and light percussion at 85 BPM. [0:45 - 1:15] Chorus: Full arrangement with lush pads, uplifting chords, and subtle strings. [1:15 - 1:45] Verse: Return to intimate piano with added guitar arpeggios. [1:45 - 2:15] Chorus: Building energy with layered harmonies and driving rhythm. [2:15 - 2:30] Outro: Fade out with piano melody alone over vinyl crackle. Genre: Lo-fi chill hop. Mood: Cozy and reflective. Instrumental only, no vocals.

31.7s

[Verse 1] Walking through the neon glow, city lights reflect below, every shadow tells a story, every corner, fading glory. [Chorus] We are the echoes in the night, burning brighter than the light, hold on tight, don't let me go, we are the echoes down below. [Verse 2] Footsteps lost on empty streets, rhythms sync to heartbeats, whispers carried by the breeze, dancing through the autumn leaves. Genre: Dreamy indie pop. Mood: Nostalgic and uplifting. Tempo: 110 BPM.

1m 4s

A bright chiptune melody in C Major, retro 8-bit video game style at 140 BPM. Instrumental only, no vocals.