口语评测（基础版）
视频服务与安全产品捉虫活动邀您参加！> HOT
文档中心>口语评测（基础版）>开发者指南>中文评测模式>拼音评测模式

拼音评测模式

最近更新时间：2024-05-30 17:12:13

我的收藏

本页目录：

评估模式描述

评测要求：支持30以内拼音。音频时长最长60秒。
评测维度：支持返回单词精准度，单词流利度；支持音素精准度。
评测功能：支持声调检测，指定发音。

请求参数

主要请求参数说明：
参数名称
类型
描述
RefText
String
被评估文本。使用拼音加上声调来表示。例如 shǎn，使用 shan3 来表示 shan（三声）
EvalMode
Integer
评估模式。8：拼音评测模式
ServerType
Integer
评估语言。1：中文
请求示例
{
    "RefText" :"shan3",
    "EvalMode": 8,
    "ServerType": 1,
    "WorkMode":1,
    "UserVoiceData":"UklGRka1CQBXQVZFZm10IBAAAAABAAEAgD4E*",
    "SeqId":1,
    "IsEnd":1,
    "VoiceFileType":3,
    "VoiceEncodeType":1,
    "SessionId":"ef5daec2-6592-11ed-9bf3-acde48001122",
    "ScoreCoeff":1.0
}

返回结果

主要返回结果说明
参数名称
类型
描述
SuggestedScore
Integer
建议评分
PronAccuracy
Integer
整体精准度
PronFluency
Integer
整体流利度
Words.PronAccuracy
Integer
单词精准度
Words.PronFluency
Integer
单词流利度
Words.MatchTag
Integer
当前词的音频与文本的匹配情况
Words.PhoneInfos.PronAccuracy
Integer
音素精准度
Words.PhoneInfos.MatchTag
Integer
当前音素的音频与文本的匹配情况
返回示例

{
  "PronAccuracy": 87.84355,
  "PronFluency": 0.92870396,
  "PronCompletion": 1,
  "Words": [
    {
      "MemBeginTime": 1080,
      "MemEndTime": 1640,
      "PronAccuracy": 87.84355,
      "PronFluency": 0.92870396,
      "Word": "shan3",
      "MatchTag": 0,
      "PhoneInfos": [
        {
          "MemBeginTime": 1080,
          "MemEndTime": 1320,
          "PronAccuracy": 92.96381,
          "DetectedStress": false,
          "Phone": "sh",
          "Stress": false,
          "ReferencePhone": "",
          "MatchTag": 0,
          "ReferenceLetter": ""
        },
        {
          "MemBeginTime": 1320,
          "MemEndTime": 1640,
          "PronAccuracy": 85.283424,
          "DetectedStress": false,
          "Phone": "an3",
          "Stress": false,
          "ReferencePhone": "",
          "MatchTag": 0,
          "ReferenceLetter": ""
        }
      ],
      "ReferenceWord": "",
      "KeywordTag": 0
    }
  ],
  "SessionId": "ef5daec2-6592-11ed-9bf3-acde48001122",
  "Status": "Finished",
  "SuggestedScore": 87.84355,
  "RefTextId": -1,
  "KeyWordHits": [],
  "UnKeyWordHits": [],
  "RequestId": "b45cfeed-f0ee-4441-bb51-c1d5c6900806"
}

声调评测

拼音评测支持声调检测，使用 {::cmd{F_TDET=true}} + 拼音 的方式进行评测。只支持4个拼音。

请求参数

主要请求参数说明
参数名称
类型
描述
RefText
String
被评估文本。使用拼音加上声调来表示。例如 shǎn，使用 shan3 来表示 shan（三声）
EvalMode
Integer
评估模式。8：拼音评测模式
ServerType
Integer
评估语言。1：中文
请求示例
{
    "RefText" :"{::cmd{F_TDET=true}}shan3",
    "EvalMode": 8,
    "ServerType": 1,
    "WorkMode":1,
    "UserVoiceData":"UklGRka1CQBXQVZFZm10IBAAAAABAAEAgD4E*",
    "SeqId":1,
    "IsEnd":1,
    "VoiceFileType":3,
    "VoiceEncodeType":1,
    "SessionId":"f64bec10-6648-11ed-96be-acde48001122",
    "ScoreCoeff":1.0
}

返回结果

主要返回结果说明
参数名称
类型
描述
SuggestedScore
Integer
建议评分
PronAccuracy
Integer
整体精准度
PronFluency
Integer
整体流利度
Words.PronAccuracy
Integer
单词精准度
Words.PronFluency
Integer
单词流利度
Words.PhoneInfos.PronAccuracy
Integer
音素精准度
Words.PhoneInfos.Phone
String
当前音频数据对应音素
Words.PhoneInfos.ReferencePhone
String
当前评估文本对应音素
返回示例
{
  "PronAccuracy": 1.3471797,
  "PronFluency": 0.9735329,
  "PronCompletion": 1,
  "Words": [
    {
      "MemBeginTime": 2660,
      "MemEndTime": 2950,
      "PronAccuracy": 1.3471797,
      "PronFluency": 0.9735329,
      "Word": "shan3",
      "MatchTag": 0,
      "PhoneInfos": [
        {
          "MemBeginTime": 2660,
          "MemEndTime": 2890,
          "PronAccuracy": 3.3700094,
          "DetectedStress": false,
          "Phone": "sh",
          "Stress": false,
          "ReferencePhone": "sh",
          "MatchTag": 0,
          "ReferenceLetter": ""
        },
        {
          "MemBeginTime": 2890,
          "MemEndTime": 2950,
          "PronAccuracy": 0.33576494,
          "DetectedStress": false,
          "Phone": "an3",
          "Stress": false,
          "ReferencePhone": "an3",
          "MatchTag": 0,
          "ReferenceLetter": ""
        }
      ],
      "ReferenceWord": "shan3",
      "KeywordTag": 0
    }
  ],
  "SessionId": "f64bec10-6648-11ed-96be-acde48001122",
  "Status": "Finished",
  "SuggestedScore": 1.3471797,
  "RefTextId": -1,
  "KeyWordHits": [],
  "UnKeyWordHits": [],
  "RequestId": "59944dce-9730-4b23-80ad-4ce38e329d06"
}

指定发音

需要在声调检测下，使用 {::cmd{F_TDET=true}} 汉字 {::pron{p1,p2..},{p3,p4..}..} 指定发音，发音单元为 拼音。

请求参数

主要请求参数说明
参数名称
类型
描述
RefText
String
被评估文本。使用| 划分多组分支
EvalMode
Integer
评估模式。8：拼音评测模式
ServerType
Integer
评估语言。1：中文
请求示例
{
    "RefText":"{::cmd{F_TDET=true}}清平乐{::pron{yue4}}"，
    "EvalMode":8,
    "ServerType":1,
    "WorkMode":1,
    "UserVoiceData":"UklGRka1CQBXQVZFZm10IBAAAAABAAEAgD4E*",
    "SeqId":1,
    "IsEnd":1,
    "VoiceFileType":3,
    "VoiceEncodeType":1,
    "SessionId":"5716-1669017375212-07754",
    "ScoreCoeff":1.0
}

返回结果

主要返回结果说明
参数名称
类型
描述
SentenceInfoSet.Words.PronAccuracy
Integer
单词精准度
SentenceInfoSet.Words.PronFluency
Integer
单词流利度
SentenceInfoSet.Words.MatchTag
Integer
当前词的音频与文本的匹配情况
SentenceInfoSet.Words.PhoneInfos.PronAccuracy
Integer
音素精准度
SentenceInfoSet.Words.PhoneInfos.MatchTag
Integer
当前音素的音频与文本的匹配情况
返回示例
{
  "PronAccuracy": 98.06444,
  "PronFluency": 0.8825564,
  "PronCompletion": 1,
  "Words": [
    {
      "MemBeginTime": 560,
      "MemEndTime": 940,
      "PronAccuracy": 97.388115,
      "PronFluency": 0.97874004,
      "Word": "清",
      "MatchTag": 0,
      "PhoneInfos": [
        {
          "MemBeginTime": 560,
          "MemEndTime": 740,
          "PronAccuracy": 98.56767,
          "DetectedStress": false,
          "Phone": "q",
          "Stress": false,
          "ReferencePhone": "q",
          "MatchTag": 0,
          "ReferenceLetter": ""
        },
        {
          "MemBeginTime": 740,
          "MemEndTime": 940,
          "PronAccuracy": 96.79834,
          "DetectedStress": false,
          "Phone": "ing1",
          "Stress": false,
          "ReferencePhone": "ing1",
          "MatchTag": 0,
          "ReferenceLetter": ""
        }
      ],
      "ReferenceWord": "qing1",
      "KeywordTag": 0,
      "Tone": {
        "Valid": false,
        "RefTone": -1,
        "HypothesisTone": -1
      }
    },
    {
      "MemBeginTime": 940,
      "MemEndTime": 1240,
      "PronAccuracy": 97.94722,
      "PronFluency": 0.987698,
      "Word": "平",
      "MatchTag": 0,
      "PhoneInfos": [
        {
          "MemBeginTime": 940,
          "MemEndTime": 1020,
          "PronAccuracy": 97.658516,
          "DetectedStress": false,
          "Phone": "p",
          "Stress": false,
          "ReferencePhone": "p",
          "MatchTag": 0,
          "ReferenceLetter": ""
        },
        {
          "MemBeginTime": 1020,
          "MemEndTime": 1240,
          "PronAccuracy": 98.09157,
          "DetectedStress": false,
          "Phone": "ing2",
          "Stress": false,
          "ReferencePhone": "ing2",
          "MatchTag": 0,
          "ReferenceLetter": ""
        }
      ],
      "ReferenceWord": "ping2",
      "KeywordTag": 0,
      "Tone": {
        "Valid": false,
        "RefTone": -1,
        "HypothesisTone": -1
      }
    },
    {
      "MemBeginTime": 1240,
      "MemEndTime": 1600,
      "PronAccuracy": 98.857956,
      "PronFluency": 0.9754167,
      "Word": "乐",
      "MatchTag": 0,
      "PhoneInfos": [
        {
          "MemBeginTime": 1240,
          "MemEndTime": 1320,
          "PronAccuracy": 99.13563,
          "DetectedStress": false,
          "Phone": "y",
          "Stress": false,
          "ReferencePhone": "y",
          "MatchTag": 0,
          "ReferenceLetter": ""
        },
        {
          "MemBeginTime": 1320,
          "MemEndTime": 1600,
          "PronAccuracy": 98.719124,
          "DetectedStress": false,
          "Phone": "ve4",
          "Stress": false,
          "ReferencePhone": "ve4",
          "MatchTag": 0,
          "ReferenceLetter": ""
        }
      ],
      "ReferenceWord": "yue4",
      "KeywordTag": 0,
      "Tone": {
        "Valid": false,
        "RefTone": -1,
        "HypothesisTone": -1
      }
    },
    {
      "MemBeginTime": 2240,
      "MemEndTime": 2610,
      "PronAccuracy": -1,
      "PronFluency": 0,
      "Word": "*",
      "MatchTag": 1,
      "PhoneInfos": [],
      "ReferenceWord": "*",
      "KeywordTag": 0,
      "Tone": {
        "Valid": false,
        "RefTone": -1,
        "HypothesisTone": -1
      }
    }
  ],
  "SessionId": "f274597a-2c44-11ee-8eef-acde48001122",
  "AudioUrl": "",
  "SentenceInfoSet": [],
  "Status": "Finished",
  "SuggestedScore": 98.06444,
  "RefTextId": -1,
  "KeyWordHits": [],
  "UnKeyWordHits": [],
  "RequestId": "920b0dc3-ff5d-4acc-b1a9-f64b6ed64355"
}


中国站
文档反馈
文档活动
message-icon联系我们
目录