You need to enable JavaScript to run this app.
导航
大模型录音文件识别闲时版
最近更新时间:2025.09.22 16:55:40首次发布时间:2025.09.22 16:55:40
复制全文
我的收藏
有用
有用
无用
无用

接口简介

本接口适用于批量/非实时场景,作业将进入闲时算力队列按调度,系统在资源空闲时自动拉起识别任务。
调用形式为提交任务(submit)轮询查询(query)

  • 优先级低于标准版/极速版,不保证实时开始;适合大体量、可容忍排队与较长完成时间的离线处理,提交任务后,任务将在24h内完成

使用限制

项目

限制说明

请求参数
音频格式/时长/大小限制

同录音文件标准版本

资源 ID

需开通 volc.bigasr.auc_idle 权限

转写结果

保证24h内可返回

接口地址

  • 提交(Submit)
    POST https://openspeech.bytedance.com/api/v3/auc/bigmodel/idle/submit

  • 查询(Query)
    POST https://openspeech.bytedance.com/api/v3/auc/bigmodel/idle/query

通用请求 Header

Key

说明

Value 示例

X-Api-App-Key

使用火山引擎控制台获取的APP ID,可参考 控制台使用FAQ-Q1

123456789x

X-Api-Access-Key

使用火山引擎控制台获取的Access Token,可参考 控制台使用FAQ-Q1

your-access-key

X-Api-Resource-Id

表示调用服务的资源信息 ID,固定值volc.bigasr.auc

volc.bigasr.auc_idle

X-Api-Request-Id

用于提交和查询任务的任务ID,推荐传入随机生成的UUID

67ee89ba-7050-4c04-a3d7-ac61a63499b3

X-Api-Sequence

发包序号,固定值,-1

-1

请求体格式(Body)

https://www.volcengine.com/docs/6561/1354868

{
  "user": {
    "uid": "你的AppID"
  },
  "audio": {
    "url": "https://example.com/audio.wav"
  },
  "request": {
    "model_name": "bigmodel"
  }
}

返回格式(成功)

提交任务响应头

服务端返回的 logid,建议用户获取和打印方便定位问题

Submit task id: cccae910-7abd-4a83-9a1e-feb91a447211
Submit task response header X-Api-Status-Code: 20000000
Submit task response header X-Api-Message: OK
Submit task response header X-Tt-Logid: 2025092212012080FF0078B8A1A25BEF2C

查询任务返回(成功)

Query task response header X-Api-Status-Code: 20000000
Query task response header X-Api-Message: OK
Query task response header X-Tt-Logid: 2025092212012080FF0078B8A1A25BEF2C
{
  "audio_info": {
    "duration": 2499
  },
  "result": {
    "additions": {
      "duration": "2499"
    },
    "text": "关闭透传。",
    "utterances": [
      {
        "end_time": 1530,
        "start_time": 450,
        "text": "关闭透传。",
        "words": [
          {
            "confidence": 0,
            "end_time": 770,
            "start_time": 450,
            "text": "关"
          },
          {
            "confidence": 0,
            "end_time": 970,
            "start_time": 770,
            "text": "闭"
          },
          {
            "confidence": 0,
            "end_time": 1210,
            "start_time": 1130,
            "text": "透"
          },
          {
            "confidence": 0,
            "end_time": 1530,
            "start_time": 1490,
            "text": "传"
          }
        ]
      }
    ]
  }
}

请求脚本

import json
import time
import uuid
import requests

def submit_task():

    submit_url = "https://openspeech.bytedance.com/api/v3/auc/bigmodel/idle/submit"

    task_id = str(uuid.uuid4())

    headers = {
        "X-Api-App-Key": appid,
        "X-Api-Access-Key": token,
        "X-Api-Resource-Id": "volc.bigasr.auc_idle",
        "X-Api-Request-Id": task_id,
        "X-Api-Sequence": "-1",
    }

    request = {
        "user": {
            "uid": "fake_uid"
        },
        "audio": {
            "url": file_url
        },
        "request": {
            "model_name": "bigmodel"
        }
    }
    print(f'Submit task id: {task_id}')
    response = requests.post(submit_url, data=json.dumps(request), headers=headers)
    if 'X-Api-Status-Code' in response.headers and response.headers["X-Api-Status-Code"] == "20000000":
        print(f'Submit task response header X-Api-Status-Code: {response.headers["X-Api-Status-Code"]}')
        print(f'Submit task response header X-Api-Message: {response.headers["X-Api-Message"]}')
        x_tt_logid = response.headers.get("X-Tt-Logid", "")
        print(f'Submit task response header X-Tt-Logid: {response.headers["X-Tt-Logid"]}\n')
        return task_id, x_tt_logid
    else:
        print(f'Submit task failed and the response headers are: {response.headers}')
        exit(1)
    return task_id

def query_task(task_id, x_tt_logid):
    query_url = "https://openspeech.bytedance.com/api/v3/auc/bigmodel/idle/query"

    headers = {
        "X-Api-App-Key": appid,
        "X-Api-Access-Key": token,
        "X-Api-Resource-Id": "volc.bigasr.auc_idle",
        "X-Api-Request-Id": task_id,
        "X-Tt-Logid": x_tt_logid, 

    }

    response = requests.post(query_url, json.dumps({}), headers=headers)

    if 'X-Api-Status-Code' in response.headers:
        print(f'Query task response header X-Api-Status-Code: {response.headers["X-Api-Status-Code"]}')
        print(f'Query task response header X-Api-Message: {response.headers["X-Api-Message"]}')
        print(f'Query task response header X-Tt-Logid: {response.headers["X-Tt-Logid"]}\n')
    else:
        print(f'Query task failed and the response headers are: {response.headers}')
        exit(1)
    return response

def main():
    task_id, x_tt_logid = submit_task()
    while True:
        query_response = query_task(task_id, x_tt_logid)
        code = query_response.headers.get('X-Api-Status-Code', "")
        if code == '20000000':  # task finished
            print(json.dumps(query_response.json(), ensure_ascii=False))
            print("SUCCESS!")
            exit(0)
        elif code != '20000001' and code != '20000002':  # task failed
            print("FAILED!")
            exit(1)
        time.sleep(5)

# 需要使用在线url,推荐使用TOS
file_url = "https://example.tos-cn-beijing.volces.com/input.wav"

# 填入控制台获取的app id和access token
appid = "xxxx"
token = "xxxx"
if __name__ == '__main__':
    main()

错误码

错误码

含义

说明

20000000

成功

20000001

正在处理中

20000002

任务在队列中

20000003

静音音频

45000001

请求参数无效

请求参数缺失必需字段 / 字段值无效 / 重复请求。

45000002

空音频

45000151

音频格式不正确

550xxxx

服务内部处理错误

55000031

服务器繁忙

服务过载,无法处理当前请求。