Support missing for model, language, and prompt fields in input_audio_transcription

[HVbajoria]

Hi team,

While working with the Azure OpenAI GPT-4o real-time transcription API, I noticed that the SDK does not yet support the model, language, and prompt fields in the input_audio_transcription configuration, despite these being documented here:
📄 https://learn.microsoft.com/en-us/azure/ai-services/openai/realtime-audio-reference

These fields are critical for:

• Selecting newer models like gpt-4o-transcribe and gpt-4o-mini-transcribe
• Providing a prompt to guide transcription behavior
• Improving accuracy and latency by specifying language (e.g. "en")

📌 I’ve opened a PR to address this gap:
🔗 #134

Tested and working with my application using the following config:

input_audio_transcription: {
  model: "gpt-4o-transcribe",
  language: "en",
  prompt: "Expect words related to a product design interview.",
}

Would love for the maintainers to review and merge this so it’s available to others using the SDK.

Thanks!