معرفی و مستندات مدل هوش مصنوعی gpt-4o-audio-preview

from openai import OpenAI
import base64
import os
client = OpenAI(
    base_url = "https://api.ai-kar.com",
    # Insert your ای آی کار (AI-KAR) key instead of <YOUR_AI-KARAPI_KEY>:
    api_key = "<YOUR_AI-KARAPI_KEY>"
)
def main():
    response = client.chat.completions.create(
        model="gpt-4o-audio-preview",
        modalities=["text", "audio"],
        audio={"voice": "alloy", "format": "wav"},
        messages=[
            {
                "role": "system",
                "content": "Speak english" # Your instructions for the model
            },
            {
                "role": "user",
                "content": "Hello" # Your question (insert it istead of Hello)
            }
        ],
        max_tokens=6000,
    )
    wav_bytes = base64.b64decode(response.choices[0].message.audio.data)
    with open("audio.wav", "wb") as f:
        f.write(wav_bytes)
    dist = os.path.abspath("audio.wav")
    print("Audio saved to:", dist)
if __name__ == "__main__":
    main()

{
  "id": "text",
  "object": "text",
  "created": 1,
  "choices": [
    {
      "index": 1,
      "message": {
        "role": "text",
        "content": "text",
        "refusal": null,
        "annotations": [
          {
            "type": "text",
            "url_citation": {
              "end_index": 1,
              "start_index": 1,
              "title": "text",
              "url": "text"
            }
          }
        ],
        "audio": {
          "id": "text",
          "data": "text",
          "transcript": "text",
          "expires_at": 1
        },
        "tool_calls": [
          {
            "id": "text",
            "type": "text",
            "function": {
              "arguments": "text",
              "name": "text"
            }
          }
        ]
      },
      "finish_reason": "stop",
      "logprobs": {
        "content": [
          {
            "bytes": [
              1
            ],
            "logprob": 1,
            "token": "text",
            "top_logprobs": [
              {
                "bytes": [
                  1
                ],
                "logprob": 1,
                "token": "text"
              }
            ]
          }
        ],
        "refusal": []
      }
    }
  ],
  "model": "text",
  "usage": {
    "prompt_tokens": 1,
    "completion_tokens": 1,
    "total_tokens": 1,
    "completion_tokens_details": {
      "accepted_prediction_tokens": 1,
      "audio_tokens": 1,
      "reasoning_tokens": 1,
      "rejected_prediction_tokens": 1
    },
    "prompt_tokens_details": {
      "audio_tokens": 1,
      "cached_tokens": 1
    }
  }
}