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

# Audio Generation

> Generate AI music, sound effects, and spoken audio from text prompts in ZeroTwo Studio.

ZeroTwo's audio Studio lets you generate AI audio — including music, sound effects, and narration — from text prompts. Describe what you need and ZeroTwo produces an audio file you can preview in-browser and download.

## Accessing audio generation

Navigate to `/studio/audio` from the **topbar** at the top of the ZeroTwo app. Click **Audio** in the topbar to open the audio workspace.

## Audio Studio vs. Voice

ZeroTwo has two separate audio features. It's important to understand the difference:

| Feature          | Audio Studio                                                              | Voice                                             |
| ---------------- | ------------------------------------------------------------------------- | ------------------------------------------------- |
| **What it does** | Generates audio files (music, sound effects, narration) from text prompts | Real-time voice conversation with AI              |
| **Output**       | Downloadable audio file (MP3, WAV, etc.)                                  | Live spoken response during a chat session        |
| **Use case**     | Background music, sound effects, produced audio assets                    | Hands-free chat, accessibility, voice interaction |
| **Location**     | `/studio/audio`                                                           | In-chat voice mode                                |

Audio Studio is for **producing audio assets**. Voice is for **speaking with AI in real time**.

## What you can create

* **Background music**: ambient tracks, genre-specific music, mood-driven compositions for videos, presentations, or apps
* **Sound effects**: UI sounds, environmental effects, specific sound descriptions
* **AI narration**: spoken audio of written text in various tones and styles
* **Jingles and short musical pieces**: branded audio, intros, outros
* **Creative audio**: experimental sound design, unique soundscapes

## How it works

<Steps>
  <Step title="Navigate to audio Studio">
    Go to `/studio/audio` via the topbar.
  </Step>

  <Step title="Describe your audio">
    Write a description of what you want: genre, mood, tempo, instruments, duration, and intended use. The more specific, the better.
  </Step>

  <Step title="Select a model">
    Choose an audio generation model. Different models are optimized for different audio types — music, voice, effects.
  </Step>

  <Step title="Set duration (if available)">
    Specify how long the audio should be, if the selected model supports duration control.
  </Step>

  <Step title="Generate and preview">
    Click Generate. When complete, preview the audio in-browser and download the file.
  </Step>
</Steps>

## Plan requirements

| Plan       | Audio generation            |
| ---------- | --------------------------- |
| Free       | Very limited or unavailable |
| Pro        | Available                   |
| Pro 2x     | Available                   |
| Plus Ultra | Unlimited                   |

<Info>
  Audio generation is primarily a Pro+ feature. Free plan users may have very limited or no access. Upgrade in **Settings → Account**.
</Info>

## Prompt essentials

Audio prompts follow different best practices depending on what you're generating:

**For background music:**
Include genre, mood, instruments, tempo, and intended duration. Also specify the context (what it's for) — this helps the model calibrate energy and style:

`Calm, professional background music for a corporate explainer video, piano and light strings, 60 seconds, no percussion, subtle and understated`

**For narration / text-to-speech:**
Provide the text you want read, plus a description of the voice — tone, pace, and style:

`Read the following text in a warm, friendly female voice at a conversational pace: "Welcome to our platform. We're glad you're here."`

**For sound effects:**
Be specific about the exact sound event, material, and environment:

`A single wooden door knock, 2 knocks, interior space with light reverb, natural sound`

**For ambient soundscapes:**
Describe the environment and the mood you want to evoke:

`Quiet office ambience: distant keyboard typing, subtle air conditioning hum, occasional muffled conversation, focused and productive atmosphere`

## Frequently asked questions

<AccordionGroup>
  <Accordion title="Is Audio Studio different from Voice mode?">
    Yes. Audio Studio generates audio files (music, effects, narration) that you download. Voice mode is real-time spoken conversation with the AI during a chat session. They are separate features.
  </Accordion>

  <Accordion title="What file formats are available?">
    MP3 and WAV are the primary formats, depending on the selected model. MP3 for general sharing; WAV for professional production. Format availability depends on the model.
  </Accordion>

  <Accordion title="Can I use generated audio commercially?">
    Generated audio from ZeroTwo Studio is generally usable for commercial purposes, subject to ZeroTwo's Terms of Service and the content policies of the underlying model providers. Review the terms for your specific use case.
  </Accordion>

  <Accordion title="Can I generate audio on the Free plan?">
    Free plan access to audio generation is very limited or unavailable. Audio is primarily a Pro+ feature. Upgrade in **Settings → Account**.
  </Accordion>
</AccordionGroup>

## Explore further

<CardGroup cols={3}>
  <Card title="Creating audio" icon="music" href="/studio/audio/create-audio">
    Step-by-step guide with prompt examples for music, effects, and narration.
  </Card>

  <Card title="Audio models" icon="cpu" href="/studio/audio/audio-models">
    Available audio generation models and their best use cases.
  </Card>

  <Card title="Troubleshooting" icon="alert-triangle" href="/studio/audio/troubleshooting">
    Fix common audio generation issues.
  </Card>
</CardGroup>
