---
title: Audio on Scenario, models and prompts
display: '*Audio* on Scenario, models and prompts'
nav_title: Audio on Scenario, models and prompts
summary:
  - "Audio is a set of separate jobs, each with its own models: music, speech, voices, effects and transcripts."
  - "Pick the job first, then the model, then write the prompt the way that model reads it: a description for music and effects, the script itself for speech."
  - "Audio is the cheapest modality, so take several; listen to each one alone and check it by transcribing it."
description: The audio jobs you can run on Scenario, where each one lives in the app, how to choose a model, and how to write a prompt for music, speech and sound effects.
section: audio
order: 10
type: concept
level: Start
roles: [creator, artist, art-director, marketer, team-lead, admin, developer]
time: { read: 10, do: 15 }
last_verified: 2026-11-11
models_checked: ["ElevenLabs v4", "Gemini 3.8 Flash TTS", "ElevenLabs Music v2.5", "Minimax Music 3.0", "Google Lyria 3.5", "ElevenLabs Sound Effects 2", "Sonilo V1.1 Text to SFX", "Sonilo V1.1 Video to SFX", "ElevenLabs Voice Changer", "Gemini 3.5 Transcribe", "ElevenLabs Sound Effects v3"]
hero: diagrams/audio-capability-map.svg
related: [generate-music, narration-and-speech, sound-effects-and-foley]
pdf: true
thumb: audio/audio-on-scenario/thumb.webp
video: tutorials/audio-on-scenario.mp4
chapters:
  - { at: 0, title: "Three kinds of audio", image: "/media/images/audio/audio-on-scenario/ch-01-three-kinds-of-audio.webp" }
  - { at: 18, title: "Open the audio panel", image: "/media/images/audio/audio-on-scenario/ch-02-open-the-audio-panel.webp" }
  - { at: 30, title: "Find a model by job", image: "/media/images/audio/audio-on-scenario/ch-03-find-a-model-by-job.webp" }
  - { at: 71, title: "The same field, renamed", image: "/media/images/audio/audio-on-scenario/ch-04-the-same-field-renamed.webp" }
  - { at: 80, title: "Write a speech take", image: "/media/images/audio/audio-on-scenario/ch-05-write-a-speech-take.webp" }
  - { at: 112.5, title: "Prompts for music and effects", image: "/media/images/audio/audio-on-scenario/ch-06-prompts-for-music-and-effects.webp" }
  - { at: 136.5, title: "Check the take", image: "/media/images/audio/audio-on-scenario/ch-07-check-the-take.webp" }
steps:
  - title: "Open the audio panel"
    text: "Create › Audio opens on a speech model, Gemini 3.8 Flash TTS, with Text, Voice, Multi-speaker and Language. Transcription is not here: it lives under Create › Text."
  - title: "Find a model by job"
    text: "Click the model name › View all models, tick Audio under Output, then one capability at a time: Speech, Music, Sound Effects or Tools. Click a model to read it, then Use this model."
  - title: "Read the field's name"
    text: "The main field is called Music Description, Sound Description or Text, depending on the model. A speech model reads Text aloud, so choose the voice in the voice setting and never describe it in the text."
  - title: "Write a speech take"
    text: "ElevenLabs v4, Public Voice Alice, Stability 0.5, Similarity Boost 0.75, Output Format MP3 44100Hz 128. Type the script in Text, with delivery tags in square brackets, and click Generate."
    prompt: |-
      [calm] Every evening at six, the keeper climbs the hundred and twelve steps of the lighthouse. [short pause] The lamp is old, the stairs are older, and the cat, as usual, is already at the top.
  - title: "Prompts for music and effects"
    text: "For music, lead with genre, mood and tempo, then the instruments, the structure and the ending; for an effect, name the source, the action and the space. End both with what to leave out."
    prompt: |-
      Short warm orchestral title cue for an animated short about a lighthouse and its cat. 90 BPM, D major. Pizzicato strings and glockenspiel, then a gentle clarinet melody. Clean final chord, no fade-out. Instrumental only, no vocals.

      A heavy wooden door with iron hinges creaks open slowly, then closes with a solid thud, inside a stone hallway with a short echo. No music, no voices.
  - title: "Check the take"
    text: "Read the length on the take's card first (here 00:12). Then Create › Text › Gemini 3.5 Transcribe, Audio › Select from Library, Generate, and compare the transcript with the script; an instrumental should transcribe to nothing."
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: "Pick the job, then the model"
    text: "Say which job you need (music, speech, a voice change, an effect or a transcript) and ask the agent to rank models for it with recommend, which is free. The brief below uses the lesson's narration line."
  - title: "Brief the agent: a speech take"
    prompt: |-
      Make one narration take with ElevenLabs v4 (model_elevenlabs-tts-v4). Read model_schema_get first for the parameter names. Public voice Alice, Stability 0.5, Similarity Boost 0.75, output format MP3 44100Hz 128. Put only the script in the text field, word for word; the tags in square brackets set the delivery and are not read aloud:
      [calm] Every evening at six, the keeper climbs the hundred and twelve steps of the lighthouse. [short pause] The lamp is old, the stairs are older, and the cat, as usual, is already at the top.

      Quote the cost with a dry run first and wait for my go.
  - title: "Check the take by transcribing it"
    prompt: |-
      Wait for the job with jobs_wait and show me the take with asset_display. Then transcribe it with Gemini 3.5 Transcribe (model_google-gemini-3-5-transcribe), default settings. Report the take's length and the transcript next to the script, word by word. Quote the cost with a dry run first and wait for my go.
  - title: "Accept or redo"
    text: "Accept when the speech matches the script word for word with the delivery you asked for and the length is what you expect. Reject missing, swapped or mispronounced words, or a tag read aloud; ask for a new take of that line only."
  - title: "File and report"
    prompt: |-
      Create a collection for this lesson with the approved take and its transcript, and report the models, the settings and every asset ID.
updated: 2026-11-11
canonical: https://learn.scenario.com/audio/audio-on-scenario
---

Last updated: 2026-11-11

# Audio on Scenario, models and prompts

> The audio jobs you can run on Scenario, where each one lives in the app, how to choose a model, and how to write a prompt for music, speech and sound effects.

In short:

- Audio is a set of separate jobs, each with its own models: music, speech, voices, effects and transcripts.
- Pick the job first, then the model, then write the prompt the way that model reads it: a description for music and effects, the script itself for speech.
- Audio is the cheapest modality, so take several; listen to each one alone and check it by transcribing it.


Audio on Scenario isn't one tool. It's a set of jobs, each with its own models: make music, speech and sound effects, change a voice you already have, and read audio back as text. This lesson shows where each job lives in the app, how to find a model by job, how to write the prompt each kind of model expects, and how to check a take. The example is one short narration line made with ElevenLabs v4 and checked with Gemini 3.5 Transcribe.

:::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.
:::

![Three groups of audio jobs: make new audio from text or video, change audio you already have, and read audio back as text](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/94fa74634c1acce652f1a3f4bdd1ef286229e3362455f5da245202d8be3ab461.svg "Make, change, read.")

:::steps Quick version
1. Open **Create** › **Audio**. The panel opens on a speech model.
2. To find a model by job, open the model menu, then **View all models**, tick **Audio**, then a capability: **Speech**, **Music**, **Sound Effects** or **Tools**.
3. Write the prompt the model reads: a description of the track for music, a description of the sound for effects, the words themselves for speech.
4. For speech, pick the voice in the voice setting, never in the text.
5. Check every take: its length first, then a transcript from **Create** › **Text** with Gemini 3.5 Transcribe.
:::

## Before you start

- **You need a project you can generate in.** Everything in this lesson runs in the normal Create panels; there's nothing to install.
- **Pick the job first, then the model.** Music, speech, voice changes, effects and transcripts are separate jobs, and a model built for one rarely does another well.
- **One field, three readings.** The audio panel has one main text field, but each kind of model reads it differently. Its name changes with the model to show you which: **Music Description**, **Sound Description** or **Text**.
- **Takes are cheap compared with video or 3D.** Plan on several takes and keep the best. The cost of each one shows on the **Generate** button before you run it.

## Step 1: Open the audio panel

Open the **Create** menu at the top of the app. Under **Generate**, two entries matter for sound:

- **Audio**: "Generate speech, music, sound effects, lip-sync, or voice cloning." Everything that makes or changes sound starts here.
- **Text**: "Transcribe audio, summarize images, or run text-to-text models." Transcription lives here, not under Audio. You'll use it in Step 6.

![The Create menu, with Audio and Text under Generate](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/23a9a46fa3634a76be66c52975c86d379415485a0318b90856285fdb742ac9d8.webp)

Click **Audio**. The panel opens on a speech model, here **Gemini 3.8 Flash TTS**, with four settings:

- **Text**: the words to speak. The placeholder reminds you that the same field takes a description for music and effects on other models.
- **Voice**: the preset voice, here **Puck (Male)**.
- **Multi-speaker**: **Add Multi-speaker** gives a second voice, for a dialogue in one take.
- **Language**: here **English (US)**.

![The audio panel on Gemini 3.8 Flash TTS, with Text, Voice, Multi-speaker, Language and Generate](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/96237d794e5295a0f54527873c9a0201a5284496fcba96c266eff7ba3b46c670.webp)

## Step 2: Find a model by job

Click the model name at the top of the panel, then **View all models**. **Choose a model** opens with every model on Scenario.

1. Under **Output**, tick **Audio**. The list narrows to the audio models.
2. Under **Capabilities**, tick one job at a time and watch the list change:
   - **Speech**: text to speech, narration and dialogue.
   - **Music**: instrumentals, cues and songs.
   - **Sound Effects**: effects from a description, and effects or music for a video.
   - **Tools**: changing a voice or cleaning up a track.
3. Click a model to read its description on the right, then **Use this model**.

Untick one capability before you tick the next, or you'll see the models that match both.

![Choose a model filtered to Audio and Speech, listing Gemini 3.8 Flash TTS, ElevenLabs v4 and other speech models](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/21481e028bfd56a76d6aa77952f8e5b04af052ef98b0d6da37874d2cb6c439cc.webp)

![The same list filtered to Music: Google Lyria 3.5, the ElevenLabs Music v2.5 models, Minimax Music 3.0 and ACE-Step 1.5](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/7160d319d51e0448370e8a201dac32c94ac8b862e5fea6714331909417eaad92.webp)

New models arrive often. Tick **New · last 30 days** under **Shortlists** to see what's new for the job you picked.

### Good starting points

| Job | Start with | Why |
|---|---|---|
| Expressive narration | ElevenLabs v4 | Audio tags such as `[whispering]`, 21 preset voices or a cloned voice, 100+ languages |
| Multilingual speech, two speakers | Gemini 3.8 Flash TTS | 30 voices, 24 languages, two speakers in one take |
| Instrumental cue with an exact length | ElevenLabs Music v2.5 | Duration from 3 to 600 seconds, instrumental switch |
| Song with lyrics | Minimax Music 3.0 or Google Lyria 3.5 | A lyrics field with section tags, or lyrics and sections in the prompt |
| Sound effect | ElevenLabs Sound Effects | Sound Effects 2 has a duration up to 30 seconds and a loop switch; v3 is the newest in the list |
| Effects for a video | Sonilo V1.1 Video to SFX | Reads the clip and returns an effects track timed to it |
| Re-voice a take | ElevenLabs Voice Changer | Keeps the words, timing and emotion |
| Transcript and subtitles | Gemini 3.5 Transcribe | Word timestamps and speaker labels |

Test two candidates on your own script or brief before you commit a series to one.

### The jobs and what each one reads

| Job | What you give it | What you get | Lesson |
|---|---|---|---|
| Music | A description of the track, sometimes lyrics, a duration | A cue or a song | [Generate music](https://learn.scenario.com/audio/generate-music.md) |
| Speech | The script itself, a voice, a language | Narration or dialogue | [Narration and speech](https://learn.scenario.com/audio/narration-and-speech.md) |
| Voice cloning | Clean recordings of one speaker | A voice you can use in speech models | [Narration and speech](https://learn.scenario.com/audio/narration-and-speech.md) |
| Voice changer | A spoken take and a target voice | The same words, timing and emotion in another voice | [Narration and speech](https://learn.scenario.com/audio/narration-and-speech.md) |
| Sound effects | A description of a source, an action and a space | A short effect, sometimes looping | [Sound for video](https://learn.scenario.com/audio/sound-effects-and-foley.md) |
| Video to audio | A silent or rough clip, an optional prompt | Effects or music timed to the picture | [Sound for video](https://learn.scenario.com/audio/sound-effects-and-foley.md) |
| Transcription | Audio or video | Text, subtitles, or a subtitled video | [Transcribe and subtitle](https://learn.scenario.com/audio/transcribe-and-subtitle.md) |
| Cleanup | Audio or video | An isolated voice, a cut, a split, extracted audio | [Sound for video](https://learn.scenario.com/audio/sound-effects-and-foley.md) |

Every audio model is also a node in [Workflows](https://learn.scenario.com/workflows/what-are-workflows.md) and runs from the [Scenario MCP](https://learn.scenario.com/mcp/what-is-scenario-mcp.md).

## Step 3: Read the field's name

Switch models in the panel and watch the main field's name change:

| Model | The field is called | It reads |
|---|---|---|
| ElevenLabs Music v2.5 | **Music Description** | A description of the track |
| ElevenLabs Sound Effects v3 | **Sound Description** | A description of a sound: source, action, space |
| ElevenLabs v4 | **Text** | The words, performed one by one |

That's the rule that saves the most takes. **The most common mistake is describing the voice in the Text field.** A speech model doesn't follow a line like this one, it reads it aloud:

```text
A calm, deep male narrator, slow and warm.
```

Choose the voice in the voice setting instead, and write only the words in **Text**.

![Three columns: music prompts name genre, tempo, instruments, structure and lyrics; speech is the script with voice, delivery tags and punctuation; sound effect prompts name source, action, material, space and duration](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/f26a070131057c7b8b85d1cb7b3bd8a2d6d500d8d53b88d21276121a33cb8e55.svg "Three prompt types, three checks.")

## Step 4: Write a speech take

Pick **ElevenLabs v4** in the model menu. The panel changes: **Text** now says "Enter any text for the model to generate speech from it (text-to-speech)", and below it are **Voice** and the voice settings.

![ElevenLabs v4 in the audio panel: Text, Voice, Public Voice, Stability, Similarity Boost, Output Format and Seed](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/e85169ea9f3c69e96b93256055b73756413348cbbb18e15fe50a16bb7bb9dbf0.webp)

1. In **Public Voice**, pick **Alice**. Leave **Voice** on **Select Model** unless you use a cloned voice.
2. Leave **Stability** at **0.5**. Lower values give a more varied, emotional reading; higher values a steadier one.
3. Leave **Similarity Boost** at **0.75** and **Output Format** at **MP3 44100Hz 128**. Set a **Seed** only if you want to repeat a take exactly.
4. Type the script in **Text**:

```text
[calm] Every evening at six, the keeper climbs the hundred and twelve steps of the lighthouse. [short pause] The lamp is old, the stairs are older, and the cat, as usual, is already at the top.
```

5. Click **Generate**.

<audio controls src="https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/2611eaed75307470b85472412cdf9ca7360d800999e27e9b9b0141bef2cdf57e.mp3"></audio>

Tags in square brackets set the delivery and are not read aloud: here `[calm]` and `[short pause]`. They work on models that support them, such as ElevenLabs v4. Punctuation sets the timing: a comma is a short breath, a period a full stop, a new paragraph a longer pause. Spell numbers and names the way they're said, as the script does with "hundred and twelve".

## Step 5: Prompts for music and effects

The same habit works for the other two jobs. For music, lead with genre, mood and tempo, then the instruments, then the structure and the ending. This is the title cue from [Generate music](https://learn.scenario.com/audio/generate-music.md):

```text
Short warm orchestral title cue for an animated short about a lighthouse and its cat. 90 BPM, D major. Pizzicato strings and glockenspiel, then a gentle clarinet melody. Clean final chord, no fade-out. Instrumental only, no vocals.
```

For an effect, name what makes the sound, what it does, and where it happens. This is the door from [Sound for video](https://learn.scenario.com/audio/sound-effects-and-foley.md):

```text
A heavy wooden door with iron hinges creaks open slowly, then closes with a solid thud, inside a stone hallway with a short echo. No music, no voices.
```

Both end by saying what to leave out. Music models tend to add vocals, and effect models tend to add music, unless you tell them not to.

## Step 6: Check the take

Audio fails quietly: a dropped word or a stray voice under a cue is easy to miss by ear. Check every take in two ways.

**First, its length.** The duration shows in the corner of the take's card in your session. This narration runs 00:12. Compare it with what you asked for, especially for music and effects with a set duration.

![The ElevenLabs v4 take in the session, 00:12 long, next to the script and the voice settings](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/7e2b24cf53cafffe018b9a24535589915939a61baad65be1cad3f6585ff8e6b5.webp)

**Then, transcribe it.**

1. Open **Create** › **Text** and pick **Gemini 3.5 Transcribe** as the model.
2. Under **Audio**, click **Select from Library** and choose the take.
3. Leave **Language hints**, **Custom vocabulary**, **Word timestamps** and **Speaker diarization** as they are for a quick check.
4. Click **Generate**, then compare the transcript with your script.

![Gemini 3.5 Transcribe with the narration loaded, Language hints, Custom vocabulary, Word timestamps and Speaker diarization](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/a59b560718b1d84966bf9c4e74396ac0cf01e0a7d7980b6e30a5d17f665da183.webp)

Speech should match the script word for word. In this take every word matches; the transcript writes "112" as digits, which is expected. An instrumental should transcribe to nothing. [Transcribe and subtitle](https://learn.scenario.com/audio/transcribe-and-subtitle.md) covers the settings in full.

:::check
**Accept when:** speech matches the script word for word with the delivery you asked for; music has the tempo, instruments and structure you asked for and no stray vocals; an effect starts and ends cleanly with no music or voice under it; the length is what you set.
**Reject when:** words are missing, swapped or mispronounced; the tags were read aloud; an instrumental contains singing or speech; the track cuts off early or fades when you asked for a clean end; a loop clicks.
:::

## Common problems

| What you see | What to do |
|---|---|
| The model reads your voice description aloud | Move the description to the voice setting and keep only the script in **Text** |
| A tag like `[calm]` is spoken | Use a model that supports tags, such as ElevenLabs v4, and keep the tag in square brackets |
| The voice is right but the reading is flat | Add a delivery tag, use shorter sentences, or lower **Stability** |
| A word is mispronounced | Spell it the way it's said and re-take that line only |
| The music ignores the tempo or structure | Put the BPM and sections earlier in the prompt, or use a model with lyrics or structure fields |
| The music has singing in it | End the prompt with "Instrumental only, no vocals", or use the model's instrumental switch |
| The effect has music under it | Add "No music, no voices", and describe a single source |
| You can't find a model | In **Choose a model**, untick the other capabilities, or search by name |

## Rights and consent

Clone or change a voice only with the speaker's permission. See [Usage restrictions and content policy](https://learn.scenario.com/browse#t-hc-accounts-legal).

## What's next

Next, [Generate music](https://learn.scenario.com/audio/generate-music.md) turns the title cue above into a finished track with an exact length and a clean ending. [Narration and speech](https://learn.scenario.com/audio/narration-and-speech.md) reuses the narration take from this lesson, and [Transcribe and subtitle](https://learn.scenario.com/audio/transcribe-and-subtitle.md) goes further with transcripts and subtitles.

Source: https://learn.scenario.com/audio/audio-on-scenario