import asyncio
import json
import protocol
import argparse

async def main():
    parser = argparse.ArgumentParser(description="ASR WebSocket Client")
    parser.add_argument("--file", type=str,
                        default="<your_file_path>",
                        help="Audio file path")
    parser.add_argument("--url", type=str, default="wss://openspeech.bytedance.com/api/v3/sauc/bigmodel_nostream",
                        help="WebSocket URL")
    parser.add_argument("--seg-duration", type=int, default=200,
                        help="Audio duration(ms) per packet, default:200")
    parser.add_argument("--api_key", type=str, default="<your_api_key>",
                        help="API key for authorization (new console auth, X-Api-Key)")
    parser.add_argument("--app_key", type=str, default="",
                        help="App key (legacy console auth, used with access_key)")
    parser.add_argument("--access_key", type=str, default="",
                        help="Access key for authorization (legacy console auth, used with app_key)")
    parser.add_argument("--resource_id", type=str, default="volc.seedasr.sauc.duration",
                        help="X-Api-Resource-Id")

    args = parser.parse_args()
    if not args.api_key and (not args.app_key or not args.access_key):
        raise ValueError("auth required: provide --api_key, or --app_key + --access_key")

    config = protocol.Config(
        api_key=args.api_key,
        app_key=args.app_key,
        access_key=args.access_key,
        resource_id=args.resource_id,
    )
    payload = {
        "audio": {
            "format": "wav",
            "codec": "raw",
            "rate": 16000,
            "bits": 16,
            "channel": 1
        },
        "request": {
            "model_name": "bigmodel",
            "enable_itn": True,
            "enable_punc": True,
            "enable_ddc": True,
            "show_utterances": True
        }
    }
    async with protocol.AsrWsClient(args.url, args.seg_duration) as client:  # 使用async with
        try:
            async for response in client.execute(args.file,config,payload):
                protocol.logger.info(f"Received response: {json.dumps(response.to_dict(), indent=2, ensure_ascii=False)}")
        except Exception as e:
            protocol.logger.error(f"ASR processing failed: {e}")

if __name__ == "__main__":
    asyncio.run(main())

{
  "code": 0,
  "event": 0,
  "is_last_package": true,
  "payload_sequence": 42,
  "payload_size": 2631,
  "payload_msg": {
    "audio_info": {
      "duration": 8040
    },
    "result": {
      "additions": {
        "log_id": "20260810152426979947089E61DF8235A3"
      },
      "text": "您好，我是您的 AI 助手。",
      "utterances": [
        {
          "additions": {
            "fixed_prefix_result": "",
            "source": "stream",
            "speaker_id": "0"
          },
          "definite": true,
          "end_time": 2000,
          "start_time": 280,
          "text": "您好，我是您的 AI 助手。",
          "words": [
            { "end_time": 360, "start_time": 280, "text": "您" },
            { "end_time": 680, "start_time": 360, "text": "好" },
            { "end_time": 880, "start_time": 680, "text": "我" },
            { "end_time": 1040, "start_time": 880, "text": "是" },
            { "end_time": 1160, "start_time": 1040, "text": "您" },
            { "end_time": 1320, "start_time": 1160, "text": "的" },
            { "end_time": 1520, "start_time": 1360, "text": "AI" },
            { "end_time": 1800, "start_time": 1600, "text": "助" },
            { "end_time": 2000, "start_time": 1800, "text": "手" }
          ]
        }
      ]
    }
  }
}