> ## Documentation Index
> Fetch the complete documentation index at: https://docs.tryflowy.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Audio Node

> Generate sound effects, music, and voiceover from a text prompt.

export const DemoVideo = ({src, poster, caption, aspectRatio = "16 / 10", className = ""}) => {
  return <figure className={`not-prose my-6 w-full ${className}`}>
      <div className="overflow-hidden rounded-2xl border border-gray-950/10 bg-gray-100 dark:border-white/10 dark:bg-white/5" style={{
    aspectRatio
  }}>
        <video src={src} poster={poster} loop autoPlay muted playsInline className="h-full w-full object-cover" />
      </div>
      {caption && <figcaption className="mt-2 text-center text-sm text-gray-500 dark:text-gray-400">
          {caption}
        </figcaption>}
    </figure>;
};

The Audio node turns text into sound — a sound effect, a music bed, a full song, or spoken narration — ready to wire into video generations and compositions.

<DemoVideo src="/videos/canvas/node-audio.mp4" poster="/images/posters/canvas/node-audio.jpg" caption="Generating a sound effect and playing it back in the node" />

## Modes

Pick a mode, describe what you want to hear, and click **Run**. The generated clip plays back right in the node as a waveform.

<Tabs>
  <Tab title="SFX">
    The default mode: a single sound effect or ambience from a description — *"helicopter takeoff: blades building, engine whine"*. Durations are `5`, `10`, `15`, `20`, or `30` seconds.

    Fine-tune with **prompt influence** (`0`–`1` — higher follows your description more literally, lower is more creative), a **loop** toggle for seamlessly looping effects, and a **fast** toggle that trades some quality for speed and unlocks the 30-second length.
  </Tab>

  <Tab title="Music">
    Three sub-types cover different jobs:

    * **Instrumental** — a music bed from genre and instrument tags (*"lofi, chill piano, soft drums, 90 BPM"*). Durations: `15`, `30`, `60`, `120`, `180`, or `240` seconds.
    * **Song** — a track with vocals. Describe the style in the prompt and structure the lyrics with `[Intro]`, `[Verse]`, `[Chorus]`, `[Bridge]`, and `[Outro]` tags — or let the lyrics optimizer write them for you.
    * **Pro Song** — for precise control: a style prompt plus timed lyrics in the lyrics editor, at a `95s` or `285s` length preset.

    Song and Pro Song both have an **instrumental** toggle to generate the track without vocals.
  </Tab>

  <Tab title="Speech">
    Narration and dialogue from a script. Pick a voice and a model tier — **Gemini Flash** for fast narration and multi-speaker dialogue, **MiniMax HD** for expressive, emotional delivery, or **ElevenLabs** for premium character voiceover — and optionally add a style direction like *"warm, slow, documentary narration"*.
  </Tab>
</Tabs>

## Connect audio downstream

Generated audio is made to be wired onward:

* **Audio → Video** — in the Video node's Default engine, a connected clip conditions the soundtrack and timing. Name it `@Audio1` in the video prompt, and connect at least one image or video alongside it.
* **Audio → Composition** — the clip lands as an audio layer on the [Composition](/canvas/nodes/composition) timeline, ready to sit under your footage.
* **Audio → Storyboard** — attach sound references while planning a sequence.

<Note>
  Audio credits scale with clip length, so longer generations cost more. See [how credits work](/billing/credits).
</Note>