Skip to main content

​
Code

from agno.agent import Agent
from agno.models.openai import OpenAIResponses
from agno.utils.audio import write_audio_to_file

agent = Agent(
    model=OpenAIResponses(
        id="gpt-5.2-audio-preview",
        modalities=["text", "audio"],
        audio={"voice": "alloy", "format": "wav"},
    ),
    debug_mode=True,
    add_history_to_context=True,
)

response_1 = agent.run("Is a golden retriever a good family dog?")
if response_1.response_audio is not None:
    write_audio_to_file(
        audio=response_1.response_audio.content, filename="tmp/answer_1.wav"
    )

response_2 = agent.run("Why do you say they are loyal?")
if response_2.response_audio is not None:
    write_audio_to_file(
        audio=response_2.response_audio.content, filename="tmp/answer_2.wav"
    )

​
Usage

1

Set up your virtual environment

uv venv --python 3.12
source .venv/bin/activate
2

Set your API key

export OPENAI_API_KEY=xxx
3

Install dependencies

uv pip install -U openai agno
4

Run Agent

python cookbook/agent_basics/multimodal/audio_multi_turn.py