文档中心 > 语音识别
语音识别

最近更新时间:2021-08-19

1 功能描述

将音频转变为特定的文本。本服务目前支持的语音类型为短语音文件,语音合成音频限制30秒之内。适用于在线对话聊天,控制口令等较短的语音识别场景。

2 请求说明

2.1 请求方式

调用地址:https://apim.unicloud.com/api/speech/asrsentence/1.0.0

请求方式:POST

返回类型:JSON

2.2 请求参数

Header如下:

参数是否必选类型可选值范围说明
AuthorizationtruestringBearer <Token>参考Token获取

Body如下:

body为二进制音频数据,如在binary中上传一个16K的音频文件,只能识别16Kbit采样率的音频。

2.3 请求示例

curl -k -H "Authorization: Bearer f38406ce-dcfe-3cdb-afb0-07b07ac5b6ad" -H "Content-Type: binary" \
   "https://apim.unicloud.com/api/speech/asrsentence/1.0.0" \
   --data-binary "@D:/tu/woaitianan.wav"  

3 返回说明

3.1 返回参数

字段类型说明
statusstring

识别结果类型:

success 完成识别

success_partial 实时识别中间结果

success_timeout音频过长,有识别结果(超过30s)

no_appkey appKey不存在

no_contextcodecontextCode不存在

no_resource appKey使用并发超限

no_license 许可不足

error 其他错误

infostringstatus为success时为识别文本,其他为空
requestIdstring唯一请求ID,之后查询调用信息时必须
detailsobject识别结果详情,包含分词等信息

3.2 返回示例

{
    "appKey": "xxxxxxx",
    "details": {
        "_hypotheses": [
            {
                "_conf": 9323,
                "_decoder": "first-path-16",
                "_items": [
                    {
                        "_id": 49,
                        "_orthography": "",
                        "_type": "terminal"
                    },
                    {
                        "_amScore": 11126,
                        "_beginTimeMs": 200,
                        "_conf": 10000,
                        "_endTimeMs": 600,
                        "_id": 198606,
                        "_lmScore": -2292,
                        "_orthography": "我爱",
                        "_phone": [
                            "ca2",
                            "chai2",
                            "chao0",
                            "bi1"
                        ],
                        "_type": "terminal"
                    },
                    {
                        "_amScore": 10305,
                        "_beginTimeMs": 600,
                        "_conf": 9687,
                        "_endTimeMs": 960,
                        "_id": 155744,
                        "_lmScore": -1278,
                        "_orthography": "北京",
                        "_phone": [
                            "ang1",
                            "chou4",
                            "bu4",
                            "cheng3"
                        ],
                        "_type": "terminal"
                    },
                    {
                        "_amScore": 23325,
                        "_beginTimeMs": 960,
                        "_conf": 9968,
                        "_endTimeMs": 1880,
                        "_id": 174700,
                        "_lmScore": -283,
                        "_orthography": "天安门",
                        "_phone": [
                            "bie4",
                            "bei3",
                            "chao0",
                            "biang2",
                            "chong1",
                            "an3"
                        ],
                        "_type": "terminal"
                    },
                    {
                        "_amScore": 11297,
                        "_beginTimeMs": 1880,
                        "_conf": 9823,
                        "_endTimeMs": 2200,
                        "_id": 174699,
                        "_lmScore": -939,
                        "_orthography": "天安",
                        "_phone": [
                            "bie4",
                            "bei3",
                            "chao0",
                            "biang2"
                        ],
                        "_type": "terminal"
                    },
                    {
                        "_amScore": 10289,
                        "_beginTimeMs": 2200,
                        "_conf": 9573,
                        "_endTimeMs": 2600,
                        "_id": 290764,
                        "_lmScore": -345,
                        "_orthography": "门上",
                        "_phone": [
                            "chong1",
                            "an3",
                            "ba0",
                            "cheng1"
                        ],
                        "_type": "terminal"
                    },
                    {
                        "_amScore": 18683,
                        "_beginTimeMs": 2600,
                        "_conf": 7487,
                        "_endTimeMs": 3520,
                        "_id": 175316,
                        "_lmScore": -533,
                        "_orthography": "太阳升",
                        "_phone": [
                            "bie4",
                            "bi1",
                            "chang2",
                            "che0",
                            "ba0",
                            "bang2"
                        ],
                        "_type": "terminal"
                    },
                    {
                        "_id": 35,
                        "_orthography": "",
                        "_type": "terminal"
                    }
                ],
                "_score": 85029,
                "_sentence": " 我爱 北京 天安门 天安 门上 太阳升 ",
                "_silence_duration": 0.0,
                "_speech_rate": 253.01
            }
        ]
    },
    "info": "我爱北京天安门,天安门上太阳升。",
    "requestId": "f687a2c9-66b3-4a3d-a012-0de8837de316",
    "status": "success"
}