---
title: Generate music
display: '*Generate* music'
nav_title: Generate music
summary:
  - "Get a cue, a bed, a loop or a song with the length, structure and vocals you asked for."
  - "Decide instrumental or vocals, exact or rough length, and where it sits before you write the prompt."
  - "Set duration in the model, not the prompt, and listen to every instrumental take alone, start to end."
description: Write a music prompt, set the length, choose instrumental or lyrics, and check the take before it goes under a picture or a voice.
section: audio
order: 20
type: guide
level: Build
roles: [creator, artist, art-director, marketer, team-lead, admin, developer]
time: { read: 9, do: 15, wait: 1 }
needs: ["Any plan with credits; the example model is ElevenLabs Music v2.5", "A brief: mood, tempo, length, where the music sits"]
last_verified: 2026-11-11
models_checked: ["ElevenLabs Music v2.5", "ElevenLabs Music Advanced v2.5", "ElevenLabs Music Extend v2.5", "Minimax Music 3.0", "Google Lyria 3.5", "Sonilo V1.1 - Text to Music", "Gemini 3.5 Transcribe"]
related: [audio-on-scenario, sound-effects-and-foley, game-trailer-cinematic]
pdf: true
run_in: [model_minimax-music-3-0, model_lyria-3-5]
thumb: audio/generate-music/thumb.webp
video: tutorials/generate-music.mp4
chapters:
  - { at: 0, title: "Three things to decide first", image: "/media/images/audio/generate-music/ch-01-three-things-to-decide-first.webp" }
  - { at: 25, title: "Pick the music model", image: "/media/images/audio/generate-music/ch-02-pick-the-music-model.webp" }
  - { at: 47, title: "Describe the track, in order", image: "/media/images/audio/generate-music/ch-03-describe-the-track-in-order.webp" }
  - { at: 58, title: "Set the length and the instrumental switch", image: "/media/images/audio/generate-music/ch-04-set-the-length-and-the-instrumental-switch.webp" }
  - { at: 74.5, title: "Generate the cue", image: "/media/images/audio/generate-music/ch-05-generate-the-cue.webp" }
  - { at: 95, title: "Check the length, the ending and the words", image: "/media/images/audio/generate-music/ch-06-check-the-length-the-ending-and-the-words.webp" }
  - { at: 121, title: "More control, with sections", image: "/media/images/audio/generate-music/ch-07-more-control-with-sections.webp" }
  - { at: 137.5, title: "A song with lyrics", image: "/media/images/audio/generate-music/ch-08-a-song-with-lyrics.webp" }
  - { at: 149, title: "Google Lyria: one prompt, with times", image: "/media/images/audio/generate-music/ch-09-google-lyria-one-prompt-with-times.webp" }
steps:
  - title: "Pick the music model"
    text: "Create › Audio › the model name › View all models, tick Audio under Output and Music under Capabilities. Pick ElevenLabs Music v2.5 and click Use this model."
  - title: "Describe the track, in order"
    text: "In Music Description: genre and mood, tempo in BPM and key, two to four lead instruments, the structure, the ending, then the vocals (here none)."
    prompt: |-
      Short warm orchestral title cue for an animated short about a lighthouse and its cat. 90 BPM, D major. Pizzicato strings and glockenspiel open for two bars, then a gentle clarinet melody over soft strings, a small swell at the midpoint, and a clean final chord on the tonic with a short natural decay. No fade-out. Instrumental only, no vocals, no choir, no spoken words.
  - title: "Set the length and the instrumental switch"
    text: "Drag Duration (s) to 20, turn on Instrumental and leave Output Format on Auto. Click Generate and listen from start to end, with headphones."
  - title: "Check the length and the ending"
    text: "The card shows the take's length: 00:20, as set. Listen for a clean final chord with a short natural decay and no fade."
  - title: "Check that it has no words"
    text: "Create › Text › Gemini 3.5 Transcribe, Audio › Select from Library › the cue, Generate. A clean instrumental returns No intelligible speech; if it returns words, turn on Instrumental and remove lines that read like lyrics."
  - title: "More control, with sections"
    text: "ElevenLabs Music Advanced v2.5 builds a track from Song Sections, each with Text, Duration (s), Styles, Negative Styles and Context Adherence; Add Song Section adds the next. ElevenLabs Music Extend v2.5 continues a take you already like."
  - title: "A song with lyrics"
    text: "Minimax Music 3.0 takes the words in Lyrics, with [Verse] and [Chorus] tags, and the sound in Style prompt. Google Lyria 3.5 takes the whole song in one Prompt with each section's time range, and up to 10 reference images."
    prompt: |-
      [Verse]
      The lamp goes on at six
      [Chorus]
      Up the stairs, the cat is there

      Warm folk song about a lighthouse keeper and his cat. [0:00 - 0:08] solo glockenspiel, soft. [0:08 - 0:30] verse with acoustic guitar. [0:30 - 0:45] chorus, full band.
agent_steps:
  - title: "Connect the Scenario MCP"
    text: "Connect your agent and sign in once at [mcp.scenario.com](https://mcp.scenario.com). Then tell it which Scenario team and project to use, so every file lands there."
  - title: "Decide the three things first"
    text: "Instrumental or vocals, exact or rough length, and where the music sits. Give them to the agent in the brief; here a 20-second instrumental title cue that ends on a clean chord for the cut."
  - title: "Brief the agent: the title cue"
    prompt: |-
      Make a title cue with ElevenLabs Music v2.5 (model_elevenlabs-music-v2-5). Read model_schema_get first for the parameter names. Duration 20 seconds, instrumental on, output format Auto. Music description, word for word:
      Short warm orchestral title cue for an animated short about a lighthouse and its cat. 90 BPM, D major. Pizzicato strings and glockenspiel open for two bars, then a gentle clarinet melody over soft strings, a small swell at the midpoint, and a clean final chord on the tonic with a short natural decay. No fade-out. Instrumental only, no vocals, no choir, no spoken words.

      Quote the cost with a dry run first and wait for my go.
  - title: "Check the length and that it has no words"
    prompt: |-
      Wait for the job with jobs_wait and show me the cue with asset_display. Report its exact length. Then transcribe it with Gemini 3.5 Transcribe (model_google-gemini-3-5-transcribe): a clean instrumental returns no speech. Quote the cost with a dry run first and wait for my go.
  - title: "Accept or redo"
    text: "Accept when the length matches the duration, the ending fits the cut, the tempo and mood match the brief and the transcript finds no speech. Reject a voice, choir or spoken word, a fade when you asked for a clean ending, or a long silence at the end."
  - title: "A song with lyrics, if you need one"
    prompt: |-
      Make the song with Google Lyria 3.5 (model_lyria-3-5), reading model_schema_get first. Prompt, word for word:
      Warm folk song about a lighthouse keeper and his cat. [0:00 - 0:08] solo glockenspiel, soft. [0:08 - 0:30] verse with acoustic guitar. [0:30 - 0:45] chorus, full band.

      Then, to compare, the same song with Minimax Music 3.0 (model_minimax-music-3-0), with these lyrics in its lyrics field, word for word, and a style prompt that names the genre, mood, instruments and voice:
      [Verse]
      The lamp goes on at six
      [Chorus]
      Up the stairs, the cat is there

      Transcribe each song with Gemini 3.5 Transcribe and check every sung word against the lyrics. Quote the cost with a dry run first and wait for my go.
  - title: "File and report"
    prompt: |-
      Create a collection for the music with the approved takes, and put each prompt in its asset's description. Report the models, the settings and every asset ID.
updated: 2026-11-11
canonical: https://learn.scenario.com/audio/generate-music
---

Last updated: 2026-11-11

# Generate music

> Write a music prompt, set the length, choose instrumental or lyrics, and check the take before it goes under a picture or a voice.

In short:

- Get a cue, a bed, a loop or a song with the length, structure and vocals you asked for.
- Decide instrumental or vocals, exact or rough length, and where it sits before you write the prompt.
- Set duration in the model, not the prompt, and listen to every instrumental take alone, start to end.


This lesson is for music you can drop into an edit: a title cue, a bed under narration, a loop, or a song with lyrics. You'll make a 20-second instrumental title cue with ElevenLabs Music v2.5, check its length and that it has no words, then see the models for sections and songs: ElevenLabs Music Advanced v2.5, Minimax Music 3.0 and Google Lyria 3.5.

:::note As of November 2026
The models, settings and screens in this article are the ones Scenario offered in November 2026. New models and updates arrive often, so check the app for newer options before you choose.
:::

:::steps Quick version
1. Decide three things: instrumental or vocals, exact or rough length, and where the music sits.
2. Open **Create** › **Audio** and pick **ElevenLabs Music v2.5** for an instrumental with an exact length.
3. In **Music Description**, describe the track in order: genre and mood, tempo, instruments, structure, ending, vocals.
4. Set **Duration (s)** and turn on **Instrumental**, then click **Generate**.
5. Check the length on the take's card, then transcribe it: a clean instrumental has no speech.
:::

## Before you start

Decide three things before you write a word of the prompt:

1. **Instrumental or vocals.** Under narration or dialogue, almost always instrumental. A song with words needs a model that takes lyrics.
2. **Exact or rough length.** A cue that ends on a cut needs an exact length, set in the model. A background bed can run long and be trimmed.
3. **Where it sits.** Under a voice, keep the arrangement sparse and out of the voice's range. As a title or a sting, it can carry the moment.

Then pick the model for the job:

| You need | Model | What it gives you |
|---|---|---|
| An instrumental with an exact length | ElevenLabs Music v2.5 | **Duration (s)** from 3 to 600 seconds and an **Instrumental** switch |
| A track built section by section | ElevenLabs Music Advanced v2.5 | Song sections, each with its own description, duration and styles |
| More of a take you like | ElevenLabs Music Extend v2.5 | Keeps the parts you want and generates new ones around them |
| A song with lyrics | Minimax Music 3.0 | A **Lyrics** field with section tags, a **Style prompt**, an **Instrumental** switch; songs of 1 to 5 minutes |
| A song in one prompt, with image references | Google Lyria 3.5 | One **Prompt** with sections and their times, up to 10 reference images, MP3 output |
| An instrumental, as an alternative | Sonilo V1.1 - Text to Music | Instrumental only |

:::flow From brief to usable take
Brief | mood, tempo, length, placement | text
Prompt | genre, BPM, instruments, ending | text
Take | duration set, instrumental on | audio
Check | length, ending, transcribe | text
:::

## Step 1: Pick the music model

Open **Create** › **Audio**, click the model name, then **View all models**. Tick **Audio** under **Output** and **Music** under **Capabilities**: the list shows the music models. Pick **ElevenLabs Music v2.5** and click **Use this model**.

![Choose a model filtered to Music, with ElevenLabs Music v2.5 highlighted among Google Lyria 3.5, the ElevenLabs Music models, Minimax Music 3.0 and ACE-Step 1.5](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/51431787e9c406c0955cf2eb806d8383bbeada4a502fd471e628cd36b27fa8ba.webp)

You can also type the model's name in the search box at the top of the model menu.

## Step 2: Describe the track, in order

The brief: a 20-second title cue for an animated short about a lighthouse and its cat, ending on a clean chord so the edit can cut on it.

The main field is now called **Music Description**. Describe the track in this order:

1. **Genre and mood:** "Short warm orchestral title cue".
2. **Tempo in BPM**, and the key if it matters: "90 BPM, D major". A number lets the editor cut on the beat.
3. **Two to four lead instruments:** pizzicato strings, glockenspiel, clarinet.
4. **The structure:** an opening, a melody, a swell at the midpoint.
5. **The ending:** a clean final chord, no fade, so the edit can cut on it.
6. **Vocals:** here, none.

```prompt
Short warm orchestral title cue for an animated short about a lighthouse and its cat. 90 BPM, D major. Pizzicato strings and glockenspiel open for two bars, then a gentle clarinet melody over soft strings, a small swell at the midpoint, and a clean final chord on the tonic with a short natural decay. No fade-out. Instrumental only, no vocals, no choir, no spoken words.
```

Say the ending you want. Without one, many music models fade out, which leaves the editor nothing to cut on.

## Step 3: Set the length and the instrumental switch

Below the description, ElevenLabs Music v2.5 opens with **Duration (s)** at 30 and **Instrumental** off.

![Music Description filled in, with Duration (s) at 30, Instrumental off, Output Format Auto and Seed](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/5fd1f528227690653b3602c38ce16f2c72dc85e972da4647dba20b1c0481d49e.webp)

1. Drag **Duration (s)** to **20**. Set the duration in the model's setting, not only in the prompt: the setting is a rule, the prompt is a hint.
2. Turn on **Instrumental**. Keep "no vocals, no choir, no spoken words" in the prompt anyway; the two together work best.
3. Leave **Output Format** on **Auto**. Set a **Seed** only to repeat a take exactly.
4. Click **Generate**, then listen from start to end, with headphones, at the volume it will play.

## Step 4: Check the length and the ending

The take appears in your session with its duration in the corner of the card: **00:20**, as set.

![The take, 00:20 long, next to the Music Description, Duration (s) at 20 and Instrumental on](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/caa313f9be799c3e22f1b7854f1f35470a29750fa1821f2b1844cb9579bee0a2.webp)

<audio controls src="https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/2a1f8dfefee98acd57e42c0baa8c9f36b2b4238038271501d537a773f2cc4b08.mp3"></audio>

Listen for the ending you asked for. Here the last chord rings out with a short natural decay and no fade, so the editor can cut on the chord or let it ring.

## Step 5: Check that it has no words

Music models sometimes sing or speak words from the prompt, and under a voice that's easy to miss. Transcribe every instrumental:

1. Open **Create** › **Text** and pick **Gemini 3.5 Transcribe**.
2. Under **Audio**, click **Select from Library** and choose the cue.
3. Click **Generate**.

On a clean instrumental, Gemini 3.5 Transcribe returns **No intelligible speech**: nothing to transcribe. That's what this cue returned. If it returns words, the model sang something from your prompt. Two fixes:

- Turn the **Instrumental** setting on, if it was off.
- Take out phrases that read like lyrics: short, rhyming or poetic lines.

[Transcribe and subtitle](https://learn.scenario.com/audio/transcribe-and-subtitle.md) explains the transcription settings.

:::check
**Accept when:** the length matches **Duration (s)**, the ending fits the cut, the tempo and mood match the brief, the transcript finds no speech in an instrumental, and every sung word in a song matches your lyrics.
**Reject when:** a voice, choir or spoken word appears in an instrumental, the track fades when you asked for a clean ending, there's a long silence at the end, or a sung line is garbled or out of order.
:::

## Step 6: More control, with sections

When one description isn't enough, **ElevenLabs Music Advanced v2.5** builds a track from sections. Under **Song Section**, each **Song Section 1**, 2 and so on has:

- **Text**: the description of that section.
- **Duration (s)**: its length.
- **Styles** and **Negative Styles**: styles to use and to avoid, added one at a time with **Add**.
- **Context Adherence**: how closely the section follows the ones around it.

Add a section with **Add Song Section**. The track plays the sections in order.

![ElevenLabs Music Advanced v2.5 with Song Section 1: Text, Duration (s) at 10, Styles, Negative Styles and Context Adherence](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/f2ff3ee5782005e0cc50de7978524bc14715fea27df398d5ce7a1b38609b4f77.webp)

**ElevenLabs Music Extend v2.5** works on a take you already like: it keeps the parts you want and continues it, or writes a new intro, bridge or ending around them.

## Step 7: A song with lyrics

**Minimax Music 3.0** has two fields:

- **Lyrics** for the words, marked by section with tags such as `[Verse]` and `[Chorus]`. The tags aren't sung.
- **Style prompt** for the sound: genre, mood, instruments, voice.

```text
[Verse]
The lamp goes on at six
[Chorus]
Up the stairs, the cat is there
```

![Minimax Music 3.0 with the lyrics in Lyrics, marked [Verse] and [Chorus], and an empty Style prompt](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/58aada6f058bf2516a1fe178a58414d35c433653bf4cb98096490148f6b3fe87.webp)

Under **Audio Settings** it also has **Instrumental**, **Auto lyrics**, **Sample rate** and **Bitrate**. Keep lines short and singable, spell words the way they're sung, and keep to one language per song.

**Google Lyria 3.5** takes the whole song in one **Prompt**, with each section and its time range, and up to 10 reference images under **Images** for the mood:

```prompt
Warm folk song about a lighthouse keeper and his cat. [0:00 - 0:08] solo glockenspiel, soft. [0:08 - 0:30] verse with acoustic guitar. [0:30 - 0:45] chorus, full band.
```

![Google Lyria 3.5 with a prompt that gives each section a time range, and the Images field below](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/41eee916955e2feec8ba8000fbe524a838a933048f8dc1c9aa6154f4089a98ce.webp)

Add images with **Drag & drop or paste**, **Select from Library**, **Upload** or **Sketch**. Keep Lyria songs to 3 minutes or less for the most reliable structure.

## Settings explained

- **Duration.** Always set it when the model has a setting. "20 seconds" in a prompt is only a hint.
- **Tempo in BPM.** Write it as a number, `90 BPM`, so the editor can cut on the beat and match later cues.
- **Structure words.** "Two-bar intro", "swell at the midpoint", "clean ending on the tonic" give the model a shape. Without them you often get a loop that fades.
- **Ending.** Ask for "no fade-out" when the picture needs a button ending. Ask for a fade when the music runs under a scene change.
- **Lyrics length.** On Minimax Music 3.0, the more lyrics you write, the longer the song.

### Loops

For a loop, such as a menu screen or a waiting state:

1. Ask for "a steady groove with no intro and no ending, the same four bars repeating".
2. Cut the loop yourself at a bar line in an audio editor.
3. Test the join by playing it twice in a row.

Effect models have a loop switch; see [Sound for video](https://learn.scenario.com/audio/sound-effects-and-foley.md).

## Common problems

| What you see | What to do |
|---|---|
| Words in an instrumental | Turn on **Instrumental**, add "no vocals, no choir, no spoken words", remove lines that read like lyrics, then transcribe again |
| The track is the wrong length | Use a model with a duration setting and set it; trim silence at the end in an editor |
| It fades out instead of ending | Ask for "a clean final chord" and "No fade-out" in the description |
| Garbled lyrics | Shorten lines, tag every section, spell words as they're sung, keep to one language |
| A generic, stock-music result | Name fewer instruments and give a structure; "cinematic, epic, emotional" alone isn't enough |
| A section tag is sung | Keep tags on their own line in square brackets, in the **Lyrics** field |
| The job failed or is stuck | See [Failed or stuck jobs](https://learn.scenario.com/browse#help-center) |

## What's next

Once a prompt gives the sound you want, keep it: save the take in a collection with the prompt in its description, or put the music node in a [workflow](https://learn.scenario.com/workflows/what-are-workflows.md) so the next episode only changes the mood line. Next, [Narration and speech](https://learn.scenario.com/audio/narration-and-speech.md) steers a voice, clones one with consent and re-voices a take.

Source: https://learn.scenario.com/audio/generate-music