# Audio

Music, spoken narration, or selecting an existing sound reference.

**Use it for:** Music, spoken narration, or selecting an existing sound reference.

**What it can do:** Switch between **Music**, **Text to speech**, and
**Select from Preview Audio**. The last option previews/reuses an existing
audio source without generating a new recording.

**Inputs:** Connected text plus the local music prompt or speech text; an
optional audio reference; a selected voice or Character voice for speech.
An audio attachment does not by itself select or clone a speech voice.

**Output:** Generated music/speech, or the selected preview audio. Use it as a
reference for compatible Video/Audio workflows or in Motion Video.

**Try music:** Describe “Gentle piano and soft strings for a sunrise harbor
scene, calm and hopeful,” review settings and displayed Gems, then generate.

**Try speech:** Choose Text to speech, select a voice, enter “A new day begins
at the harbor,” then review the displayed Gems and generate.

**Remember:** Preview Audio reuses a file. Text to speech needs a voice.
Music references must finish generating/uploading before they can be used.
Use the price shown in the app rather than a fixed tutorial price.

### Watch the setup

These short demos show setup or existing results. Generation and analysis are
started separately after you review the inputs and displayed price or allowance.

#### Music prompt and settings

![Set up a music prompt and audio options. Setup only; no generation.](https://docs-media.letsgen.app/canvas/audio/audio-music/2026-10-07-fe25ce8313dffff9.png)

[Watch video: Set up a music prompt and audio options. Setup only; no generation.](https://docs-media.letsgen.app/canvas/audio/audio-music/2026-10-07-aea7f63d06345e01.mp4)

#### Text to speech setup

![Choose Text to speech and enter the spoken words. Setup only; no generation.](https://docs-media.letsgen.app/canvas/audio/audio-speech/2026-10-07-02f2d57495cbe404.png)

[Watch video: Choose Text to speech and enter the spoken words. Setup only; no generation.](https://docs-media.letsgen.app/canvas/audio/audio-speech/2026-10-07-5361a3e1e7807845.mp4)