Any plans to support base64 audio input for gpt-audio in AzureChatOpenAI?

[JunHyeongLee92]

Hello LangChain team,

Recently, OpenAI introduced the gpt-audio model, which allows direct audio input processing via the Chat Completions API.

In our use case, we would like to use the gpt-audio model in an Azure environment to directly input audio files (MP3) and have the model process them.
I’m wondering if there are any plans to update AzureChatOpenAI to support base64 audio input for gpt-audio.

If such support is planned, could you please share the expected implementation approach (e.g., allowing HumanMessage to accept base64 audio data) and any estimated release timeline?

Thank you.