文档反馈官招募中，报名立赚积分兑换代金券！> HOT
文档中心>口语评测（新版）>开发者指南>英文评测模式>单词纠错评测模式

单词纠错评测模式

最近更新时间：2025-08-15 15:21:22

我的收藏

本页目录：

评测模式描述

评测要求：支持单词，组合词评测。音频数据最长60秒。
评测维度：支持返回单词精准度，单词流利度；支持返回音素精准度。
评测功能：支持重音检测，标准音素，音素到字母映射，音素到国际音标转换，指定发音，指定国际音标。

请求参数

主要请求参数说明：
参数名称
类型
描述
ref_text
String
被评估文本。支持单词，组合词，缩写
eval_mode
Integer
评估模式。4：单词纠错模式
请求示例

# 参数示例为websocket连接URL展开, 如：soe.cloud.tencent.com/soe/api/1306***?eval_mode=0&voice_format=1&...
server_engine_type=16k_en
eval_mode=4
rec_mode=1
ref_text"happy"
score_coeff=1.000000
voice_format=1

返回结果

主要返回结果说明：
参数名称
类型
描述
SuggestedScore
Float
建议评分。一分制
PronAccuracy
Float
整体精准度。一分制
PronFluency
Float
整体流利度
Words.PronAccuracy
Float
单词精准度。百分制
Words.PronFluency
Float
单词流利度
Words.PhoneInfos.PronAccuracy
Float
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
Words.PhoneInfos.Stress
Boolean
当前评估文本对应音素是否为重音
Words.PhoneInfos.DetectedStress
Boolean
当前音频数据对应音素是否为重音
返回示例
{
    "code": 0,
    "message": "f9376259-5e0f-4de9-96f9-681d1634c281_1",
    "voice_id": "f9376259-5e0f-4de9-96f9-681d1634c281",
    "result": {
        "SuggestedScore": 0.973709225654602,
        "PronAccuracy": 0.973709225654602,
        "PronFluency": 0.9604969024658203,
        "PronCompletion": 1,
        "Words": [
            {
                "MemBeginTime": 200,
                "MemEndTime": 650,
                "PronAccuracy": 97.37092590332031,
                "PronFluency": 0.9604969024658203,
                "ReferenceWord": "",
                "Word": "happy",
                "MatchTag": 0,
                "KeywordTag": 0,
                "PhoneInfos": [
                    {
                        "MemBeginTime": 200,
                        "MemEndTime": 260,
                        "PronAccuracy": 0.9145820140838623,
                        "DetectedStress": true,
                        "Phone": "hh",
                        "ReferencePhone": "hh",
                        "ReferenceLetter": "",
                        "Stress": true,
                        "MatchTag": 0
                    },
                    {
                        "MemBeginTime": 260,
                        "MemEndTime": 340,
                        "PronAccuracy": 0.9899160861968994,
                        "DetectedStress": false,
                        "Phone": "ae",
                        "ReferencePhone": "ae",
                        "ReferenceLetter": "",
                        "Stress": false,
                        "MatchTag": 0
                    },
                    {
                        "MemBeginTime": 340,
                        "MemEndTime": 420,
                        "PronAccuracy": 0.9958605766296387,
                        "DetectedStress": false,
                        "Phone": "p",
                        "ReferencePhone": "p",
                        "ReferenceLetter": "",
                        "Stress": false,
                        "MatchTag": 0
                    },
                    {
                        "MemBeginTime": 420,
                        "MemEndTime": 650,
                        "PronAccuracy": 0.9944781064987183,
                        "DetectedStress": false,
                        "Phone": "iy",
                        "ReferencePhone": "iy",
                        "ReferenceLetter": "",
                        "Stress": false,
                        "MatchTag": 0
                    }
                ],
                "Tone": {
                    "Valid": false,
                    "RefTone": -1,
                    "HypothesisTone": -1
                }
            }
        ],
        "SentenceId": -1,
        "RefTextId": -1,
        "KeyWordHits": null,
        "UnKeyWordHits": null
    },
    "final": 1
}

音素到字母的映射

通过该功能，可以对音素映射字母进行标记。音素到字母映射结构：{::cmd{F_P2L=true}} + 评估文本。

请求参数

主要请求参数说明：
参数名称
类型
描述
ref_text
String
被评估文本
eval_mode
Integer
评估模式。4：单词纠错模式
请求示例

# 参数示例为websocket连接URL展开, 如：soe.cloud.tencent.com/soe/api/1306***?eval_mode=0&voice_format=1&...
server_engine_type=16k_en
eval_mode=4
rec_mode=1
ref_text"{::cmd{F_P2L=true}} happy"
score_coeff=1.000000
voice_format=1

返回结果

主要返回结果：
参数名称
类型
描述
Words.Word
String
当前单词
Words.PhoneInfos.Phone
String
当前单词的音素
Words.PhoneInfos.ReferenceLetter
String
当前单词的音素映射的字母
返回示例
{
    "code": 0,
    "message": "61a44c5f-b7f9-43d7-9626-dc0b87f8b612_1",
    "voice_id": "61a44c5f-b7f9-43d7-9626-dc0b87f8b612",
    "result": {
        "SuggestedScore": 0.973709225654602,
        "PronAccuracy": 0.973709225654602,
        "PronFluency": 0.9604969024658203,
        "PronCompletion": 1,
        "Words": [
            {
                "MemBeginTime": 200,
                "MemEndTime": 650,
                "PronAccuracy": 97.37092590332031,
                "PronFluency": 0.9604969024658203,
                "ReferenceWord": "",
                "Word": "happy",
                "MatchTag": 0,
                "KeywordTag": 0,
                "PhoneInfos": [
                    {
                        "MemBeginTime": 200,
                        "MemEndTime": 260,
                        "PronAccuracy": 0.9145820140838623,
                        "DetectedStress": true,
                        "Phone": "hh",
                        "ReferencePhone": "hh",
                        "ReferenceLetter": "h_0",
                        "Stress": true,
                        "MatchTag": 0
                    },
                    {
                        "MemBeginTime": 260,
                        "MemEndTime": 340,
                        "PronAccuracy": 0.9899160861968994,
                        "DetectedStress": false,
                        "Phone": "ae",
                        "ReferencePhone": "ae",
                        "ReferenceLetter": "a_1",
                        "Stress": false,
                        "MatchTag": 0
                    },
                    {
                        "MemBeginTime": 340,
                        "MemEndTime": 420,
                        "PronAccuracy": 0.9958605766296387,
                        "DetectedStress": false,
                        "Phone": "p",
                        "ReferencePhone": "p",
                        "ReferenceLetter": "pp_2",
                        "Stress": false,
                        "MatchTag": 0
                    },
                    {
                        "MemBeginTime": 420,
                        "MemEndTime": 650,
                        "PronAccuracy": 0.9944781064987183,
                        "DetectedStress": false,
                        "Phone": "iy",
                        "ReferencePhone": "iy",
                        "ReferenceLetter": "y_4",
                        "Stress": false,
                        "MatchTag": 0
                    }
                ],
                "Tone": {
                    "Valid": false,
                    "RefTone": -1,
                    "HypothesisTone": -1
                }
            }
        ],
        "SentenceId": -1,
        "RefTextId": -1,
        "KeyWordHits": null,
        "UnKeyWordHits": null
    },
    "final": 1
}

音素到国际音标转换

默认返回智聆音素，使用 {::cmd{F_IPA=true}} + 单词 开启⾳素到国际⾳标转换功能。

请求参数

主要请求参数说明：
参数名称
类型
描述
ref_text
String
被评估文本
eval_mode
Integer
评估模式。4：单词纠错评测模式
请求示例

# 参数示例为websocket连接URL展开, 如：soe.cloud.tencent.com/soe/api/1306***?eval_mode=0&voice_format=1&...
server_engine_type=16k_en
eval_mode=4
rec_mode=1
ref_text"{::cmd{F_IPA=true}} happy"
score_coeff=1.000000
voice_format=1

返回结果

返回结果说明：
参数名称
类型
描述
Words.PhoneInfos.Phone
String
当前音节
返回示例
{
    "code": 0,
    "message": "2d21cbe7-01c1-4457-aa0c-cc5fb85b27af_1",
    "voice_id": "2d21cbe7-01c1-4457-aa0c-cc5fb85b27af",
    "result": {
        "SuggestedScore": 0.973709225654602,
        "PronAccuracy": 0.973709225654602,
        "PronFluency": 0.9604969024658203,
        "PronCompletion": 1,
        "Words": [
            {
                "MemBeginTime": 200,
                "MemEndTime": 650,
                "PronAccuracy": 97.37092590332031,
                "PronFluency": 0.9604969024658203,
                "ReferenceWord": "",
                "Word": "happy",
                "MatchTag": 0,
                "KeywordTag": 0,
                "PhoneInfos": [
                    {
                        "MemBeginTime": 200,
                        "MemEndTime": 260,
                        "PronAccuracy": 0.9145820140838623,
                        "DetectedStress": true,
                        "Phone": "'h",
                        "ReferencePhone": "'h",
                        "ReferenceLetter": "",
                        "Stress": true,
                        "MatchTag": 0
                    },
                    {
                        "MemBeginTime": 260,
                        "MemEndTime": 340,
                        "PronAccuracy": 0.9899160861968994,
                        "DetectedStress": false,
                        "Phone": "æ",
                        "ReferencePhone": "æ",
                        "ReferenceLetter": "",
                        "Stress": false,
                        "MatchTag": 0
                    },
                    {
                        "MemBeginTime": 340,
                        "MemEndTime": 420,
                        "PronAccuracy": 0.9958605766296387,
                        "DetectedStress": false,
                        "Phone": "p",
                        "ReferencePhone": "p",
                        "ReferenceLetter": "",
                        "Stress": false,
                        "MatchTag": 0
                    },
                    {
                        "MemBeginTime": 420,
                        "MemEndTime": 650,
                        "PronAccuracy": 0.9944781064987183,
                        "DetectedStress": false,
                        "Phone": "i",
                        "ReferencePhone": "i",
                        "ReferenceLetter": "",
                        "Stress": false,
                        "MatchTag": 0
                    }
                ],
                "Tone": {
                    "Valid": false,
                    "RefTone": -1,
                    "HypothesisTone": -1
                }
            }
        ],
        "SentenceId": -1,
        "RefTextId": -1,
        "KeyWordHits": null,
        "UnKeyWordHits": null
    },
    "final": 1
}


指定发音

使用 单词{::pron{p1,p2..},{p3,p4..}..} 指定发音，需要将国际音标通过音素映射表转换为智聆音素，参考音素映射表 -- 智聆音素。

请求参数

主要请求参数说明：
参数名称
类型
描述
ref_text
String
被评估文本
eval_mode
Integer
评估模式。4：单词纠错评测模式
请求示例
音标
单词
多音词
# 参数示例为websocket连接URL展开, 如：soe.cloud.tencent.com/soe/api/1306***?eval_mode=0&voice_format=1&...
server_engine_type=16k_en
eval_mode=4
rec_mode=1
ref_text"au{::pron{aw}}"
score_coeff=1.000000
voice_format=1
server_engine_type=16k_en
eval_mode=4
rec_mode=1
ref_text"happy{::pron{hh,ae,p,iy}}"
score_coeff=1.000000
voice_format=1
server_engine_type=16k_en
eval_mode=4
rec_mode=1
ref_text"use{::pron{y,uw,z},{y,uw,s}}"
score_coeff=1.000000
voice_format=1

返回结果

主要返回结果说明：
参数名称
类型
描述
SuggestedScore
Float
建议评分
PronAccuracy
Float
整体精准度
PronFluency
Float
整体流利度
Words.PronAccuracy
Float
单词精准度
Words.PronFluency
Float
单词流利度
Words.PhoneInfos.PronAccuracy
Float
音素精准度
返回示例
{
    "code": 0,
    "message": "45bca24a-29c0-4a7e-95b9-02a6c69172a6_1",
    "voice_id": "45bca24a-29c0-4a7e-95b9-02a6c69172a6",
    "result": {
        "SuggestedScore": 0.973709225654602,
        "PronAccuracy": 0.973709225654602,
        "PronFluency": 0.9604969024658203,
        "PronCompletion": 1,
        "Words": [
            {
                "MemBeginTime": 200,
                "MemEndTime": 650,
                "PronAccuracy": 97.37092590332031,
                "PronFluency": 0.9604969024658203,
                "ReferenceWord": "",
                "Word": "happy",
                "MatchTag": 0,
                "KeywordTag": 0,
                "PhoneInfos": [
                    {
                        "MemBeginTime": 200,
                        "MemEndTime": 260,
                        "PronAccuracy": 0.9145820140838623,
                        "DetectedStress": true,
                        "Phone": "hh",
                        "ReferencePhone": "hh",
                        "ReferenceLetter": "",
                        "Stress": true,
                        "MatchTag": 0
                    },
                    {
                        "MemBeginTime": 260,
                        "MemEndTime": 340,
                        "PronAccuracy": 0.9899160861968994,
                        "DetectedStress": false,
                        "Phone": "ae",
                        "ReferencePhone": "ae",
                        "ReferenceLetter": "",
                        "Stress": false,
                        "MatchTag": 0
                    },
                    {
                        "MemBeginTime": 340,
                        "MemEndTime": 420,
                        "PronAccuracy": 0.9958605766296387,
                        "DetectedStress": false,
                        "Phone": "p",
                        "ReferencePhone": "p",
                        "ReferenceLetter": "",
                        "Stress": false,
                        "MatchTag": 0
                    },
                    {
                        "MemBeginTime": 420,
                        "MemEndTime": 650,
                        "PronAccuracy": 0.9944781064987183,
                        "DetectedStress": false,
                        "Phone": "iy",
                        "ReferencePhone": "iy",
                        "ReferenceLetter": "",
                        "Stress": false,
                        "MatchTag": 0
                    }
                ],
                "Tone": {
                    "Valid": false,
                    "RefTone": -1,
                    "HypothesisTone": -1
                }
            }
        ],
        "SentenceId": -1,
        "RefTextId": -1,
        "KeyWordHits": null,
        "UnKeyWordHits": null
    },
    "final": 1
}

指定国际音标

使用 单词{::ipapron{p1,p2..},{p3,p4..}..} 指定国际音标参考，音素映射表 -- 国际音标。默认返回智聆音素，返回国际音标需要使用音素到国际音标转换。

请求参数

主要请求参数说明：
参数名称
类型
描述
ref_text
String
被评估文本
eval_mode
Integer
评估模式。4：单词纠错评测模式
请求示例
音标
单词
多音词
# 参数示例为websocket连接URL展开, 如：soe.cloud.tencent.com/soe/api/1306***?eval_mode=0&voice_format=1&...
server_engine_type=16k_en
eval_mode=4
rec_mode=1
ref_text"au{::ipapron{aʊ}}"
score_coeff=1.000000
voice_format=1
server_engine_type=16k_en
eval_mode=4
rec_mode=1
ref_text"happy{::ipapron{'h,æ,p,i}}"
score_coeff=1.000000
voice_format=1
server_engine_type=16k_en
eval_mode=4
rec_mode=1
ref_text"use{::ipapron{j,u,z},{j,u,s}}"
score_coeff=1.000000
voice_format=1

返回结果

主要返回结果说明：
参数名称
类型
描述
SuggestedScore
Float
建议评分
PronAccuracy
Float
整体精准度
PronFluency
Float
整体流利度
Words.PronAccuracy
Float
单词精准度
Words.PronFluency
Float
单词流利度
Words.PhoneInfos.PronAccuracy
Float
音素精准度
返回示例
{
    "code": 0,
    "message": "fdde8b88-d229-435b-8fc1-b624e42ec65d_1",
    "voice_id": "fdde8b88-d229-435b-8fc1-b624e42ec65d",
    "result": {
        "SuggestedScore": 0.973709225654602,
        "PronAccuracy": 0.973709225654602,
        "PronFluency": 0.9604969024658203,
        "PronCompletion": 1,
        "Words": [
            {
                "MemBeginTime": 200,
                "MemEndTime": 650,
                "PronAccuracy": 97.37092590332031,
                "PronFluency": 0.9604969024658203,
                "ReferenceWord": "",
                "Word": "happy",
                "MatchTag": 0,
                "KeywordTag": 0,
                "PhoneInfos": [
                    {
                        "MemBeginTime": 200,
                        "MemEndTime": 260,
                        "PronAccuracy": 0.9145820140838623,
                        "DetectedStress": true,
                        "Phone": "hh",
                        "ReferencePhone": "hh",
                        "ReferenceLetter": "",
                        "Stress": true,
                        "MatchTag": 0
                    },
                    {
                        "MemBeginTime": 260,
                        "MemEndTime": 340,
                        "PronAccuracy": 0.9899160861968994,
                        "DetectedStress": false,
                        "Phone": "ae",
                        "ReferencePhone": "ae",
                        "ReferenceLetter": "",
                        "Stress": false,
                        "MatchTag": 0
                    },
                    {
                        "MemBeginTime": 340,
                        "MemEndTime": 420,
                        "PronAccuracy": 0.9958605766296387,
                        "DetectedStress": false,
                        "Phone": "p",
                        "ReferencePhone": "p",
                        "ReferenceLetter": "",
                        "Stress": false,
                        "MatchTag": 0
                    },
                    {
                        "MemBeginTime": 420,
                        "MemEndTime": 650,
                        "PronAccuracy": 0.9944781064987183,
                        "DetectedStress": false,
                        "Phone": "iy",
                        "ReferencePhone": "iy",
                        "ReferenceLetter": "",
                        "Stress": false,
                        "MatchTag": 0
                    }
                ],
                "Tone": {
                    "Valid": false,
                    "RefTone": -1,
                    "HypothesisTone": -1
                }
            }
        ],
        "SentenceId": -1,
        "RefTextId": -1,
        "KeyWordHits": null,
        "UnKeyWordHits": null
    },
    "final": 1
}

中国站
目录