OpenAI 兼容语音识别

from openai import OpenAI
import os

try:
  client = OpenAI(
    # 如果您尚未配置环境变量，请将以下代码行替换为您的 API Key：api_key = "sk-xxx",
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    base_url="https://dashscope.aliyuncs.com/compatible-mode/v1",
  )
  

  stream_enabled = False  # 是否启用流式输出
  completion = client.chat.completions.create(
    model="qwen3-asr-flash",
    messages=[
      {
        "content": [
          {
            "type": "input_audio",
            "input_audio": {
              "data": "https://dashscope.oss-cn-beijing.aliyuncs.com/audios/welcome.mp3"
            }
          }
        ],
        "role": "user"
      }
    ],
    stream=stream_enabled,
    # 当 stream 设置为 False 时，不能设置 stream_options 参数
    # stream_options={"include_usage": True},
    extra_body={
      "asr_options": {
        # "language": "zh",
        "enable_itn": False
      }
    }
  )
  if stream_enabled:
    full_content = ""
    print("流式输出内容为：")
    for chunk in completion:
      # 若 stream_options.include_usage 为 True，最后一个 chunk 的 choices 字段为空列表，应跳过（可通过 chunk.usage 获取 token 用量）
      print(chunk)
      if chunk.choices and chunk.choices[0].delta.content:
        full_content += chunk.choices[0].delta.content
    print(f"完整内容为：{full_content}")
  else:
    print(f"非流式输出内容为：{completion.choices[0].message.content}")
except Exception as e:
  print(f"错误信息：{e}")