> ## Documentation Index
> Fetch the complete documentation index at: https://agno-v2-codex-docs-audit-20260719-0149.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Audio Generation Tools

> Generate audio with text-to-speech tools in Agno agents.

The following example demonstrates how to generate audio using the ElevenLabs tool with an agent. See [Eleven Labs](https://elevenlabs.io/) for more details.

## Prerequisites

You need to install the `elevenlabs` library and an API key which can be obtained from [Eleven Labs](https://elevenlabs.io/)

```bash theme={null}
uv pip install -U agno elevenlabs google-genai
```

Set the `ELEVEN_LABS_API_KEY` and `GOOGLE_API_KEY` environment variables.

```bash theme={null}
export ELEVEN_LABS_API_KEY=****
export GOOGLE_API_KEY=****
```

```python audio_agent.py theme={null}
import base64
from textwrap import dedent

from agno.agent import Agent
from agno.models.google import Gemini
from agno.tools.eleven_labs import ElevenLabsTools
from agno.utils.media import save_base64_data

audio_agent = Agent(
    model=Gemini(id="gemini-2.5-pro"),
    tools=[
        ElevenLabsTools(
            voice_id="21m00Tcm4TlvDq8ikWAM",
            model_id="eleven_multilingual_v2",
            target_directory="audio_generations",
        )
    ],
    description="You are an AI agent that can generate audio using the ElevenLabs API.",
    instructions=[
        dedent(
            """
            You have access to the ElevenLabs toolkit:
            - Use the `text_to_speech` tool to convert text or speech content into natural voice audio.
            - Use the `generate_sound_effect` tool to create sound effects from text descriptions.
            Keep the audio prompt as defined by the user.
            """
        ),
    ],
    markdown=True,
)

response = audio_agent.run(
    "Generate a very long audio of history of french revolution and tell me which subject it belongs to.",
    debug_mode=True,
)

if response.audio:
    print("Agent response:", response.content)
    base64_audio = base64.b64encode(response.audio[0].content).decode("utf-8")
    save_base64_data(base64_audio, "tmp/french_revolution.mp3")
    print("Successfully saved generated speech to tmp/french_revolution.mp3")


audio_agent.print_response("Generate a kick sound effect")
```

## Run Agent

```bash theme={null}
python audio_agent.py
```