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

# OpenAI Audio Input And Output Multi Turn

> Send audio input to gpt-audio and carry the conversation across turns with audio replies.

```python audio_input_and_output_multi_turn.py theme={null}
"""
Openai Audio Input And Output Multi Turn
========================================

Cookbook example for `openai/chat/audio_input_and_output_multi_turn.py`.
"""

from pathlib import Path

import requests
from agno.agent import Agent, RunOutput  # noqa
from agno.media import Audio
from agno.models.openai import OpenAIChat
from agno.utils.audio import write_audio_to_file

# ---------------------------------------------------------------------------
# Create Agent
# ---------------------------------------------------------------------------

# Fetch the audio file and convert it to a base64 encoded string
url = "https://openaiassets.blob.core.windows.net/$web/API/docs/audio/alloy.wav"
response = requests.get(url)
response.raise_for_status()
wav_data = response.content

# Provide the agent with the audio file and audio configuration and get result as text + audio
agent = Agent(
    model=OpenAIChat(
        id="gpt-audio",
        modalities=["text", "audio"],
        audio={"voice": "alloy", "format": "wav"},
    ),
    # Set add_history_to_context=true to add the previous chat history to the context sent to the Model.
    add_history_to_context=True,
    # Number of historical responses to add to the messages.
    num_history_runs=3,
)
run_output: RunOutput = agent.run(
    input="What is in this audio?", audio=[Audio(content=wav_data, format="wav")]
)

filename = Path(__file__).parent.joinpath("tmp/conversation_response_1.wav")
filename.unlink(missing_ok=True)
filename.parent.mkdir(parents=True, exist_ok=True)

# Save the response audio to a file
if run_output.response_audio is not None:
    write_audio_to_file(audio=run_output.response_audio.content, filename=str(filename))

run_output: RunOutput = agent.run("Tell me something more about the audio")

filename = Path(__file__).parent.joinpath("tmp/conversation_response_2.wav")
filename.unlink(missing_ok=True)

# Save the response audio to a file
if run_output.response_audio is not None:
    write_audio_to_file(audio=run_output.response_audio.content, filename=str(filename))

run_output: RunOutput = agent.run("Now tell me a 5 second story")

filename = Path(__file__).parent.joinpath("tmp/conversation_response_3.wav")
filename.unlink(missing_ok=True)

# Save the response audio to a file
if run_output.response_audio is not None:
    write_audio_to_file(audio=run_output.response_audio.content, filename=str(filename))

# ---------------------------------------------------------------------------
# Run Agent
# ---------------------------------------------------------------------------

if __name__ == "__main__":
    pass
```

## Run the Example

<Snippet file="create-venv-step.mdx" />

**1. Install dependencies**

```bash theme={null}
uv pip install -U agno openai requests
```

**2. Export your OpenAI API key**

```bash Mac/Linux theme={null}
export OPENAI_API_KEY="your_openai_api_key_here"
```

```bash Windows theme={null}
$Env:OPENAI_API_KEY="your_openai_api_key_here"
```

**3. Run the example**

Save the code above as `audio_input_and_output_multi_turn.py`, then run:

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

Full source: [cookbook/90\_models/openai/chat/audio\_input\_and\_output\_multi\_turn.py](https://github.com/agno-agi/agno/blob/v3.0.4/cookbook/90_models/openai/chat/audio_input_and_output_multi_turn.py)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.