Audio API Resource is Missing Timeout Handling

[TedTimbrell]

Describe the feature or improvement you're requesting

Audio transcription can be time intensive (and in my case also somewhat time critical). In certain cases the OpenAI API is taking 20+ seconds to process audio and with the currently library we have no way of timing out the call, potentially leaving a worker or thread hanging until resolved ungracefully (such as an overall network response timeout or worker timeout). As is this forces users to either reimplement the call in requests to get access to network timeouts or re-implement the timeout logic on top of this method call via threads or subprocesses.

Additional context

No response

[logankilpatrick]

Verified this, checking with the team. The Audio class inherits from APIResource which has the request_timeout option so not sure what is going on here.

[RobertCraigie]

This is fixed in our v1 beta as we've unified to a single timeout argument which is passed to the client constructor and can be configured per-request using client.with_options.

from openai import OpenAI

# Configure the default for all requests:
client = OpenAI(
    # default is 60s
    timeout=20.0,
)

# More granular control:
client = OpenAI(
    timeout=httpx.Timeout(60.0, read=5.0, write=10.0, connect=2.0),
)

# Override per-request:
client.with_options(timeout=5 * 1000).chat.completions.create(
    model="gpt-3.5-turbo",
    messages=[
        {
            "role": "user",
            "content": "How can I list all files in a directory using Python?",
        }
    ],
)