tencentcloud-sdk-asr 3.0.1161 → 3.0.1184

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 3d44e052c1c113ad1b936064dfe2bc51dbb54652
-  data.tar.gz: 151be8386b0ba31622eefda3c8f95a762528d579
+  metadata.gz: 182a61e43aa111f27e93c531075aea02d2fe06d8
+  data.tar.gz: 474059e7f927a31724cc8e0f5ae4326f24ddd43d
 SHA512:
-  metadata.gz: 5c6b40c1d1d23303c1f41e9a5daf3f67f23be6b31a6ce3ad645c28c066830678fed4b3ca22f7692df016fd2ed730e71863501c8a29a9dfe76fb009f587029b85
-  data.tar.gz: 5df0f4a26e00a3fec02410a88dba65f35452afe4ff3b6ca9c7575e4a9718194ba595accfc562d1a7e6d45ad43ae0f9790d964f686360a91995feb4549eb3acaf
+  metadata.gz: 977cb2d4e3566d45340174af95a0e92cc394c069c7cc72f3da5657c61d87bacabfa7f43bf3c1182ff1f3a3d21b677263d0332817c26d288548248f643fa92cb0
+  data.tar.gz: 2cac0703665a4af4c1c47ffe38ac0ad3f1075eb8cdbdcaa6fdbafb5d19964b480e68288635be41409a3da3fa8eba83f6669412dd98be27d1d90a991ca4f02ac3

data/lib/VERSION

@@ -1 +1 @@
-3.0.1161
+3.0.1184

data/lib/v20190614/models.rb

@@ -305,8 +305,8 @@ module TencentCloud
 
         attr_accessor :ModelName, :TextUrl, :ModelType, :TagInfos
         extend Gem::Deprecate
-        deprecate :TagInfos, :none, 2025, 11
-        deprecate :TagInfos=, :none, 2025, 11
+        deprecate :TagInfos, :none, 2025, 12
+        deprecate :TagInfos=, :none, 2025, 12
 
         def initialize(modelname=nil, texturl=nil, modeltype=nil, taginfos=nil)
           @ModelName = modelname
@@ -352,13 +352,13 @@ module TencentCloud
         # **注意：电话通讯场景，请务必使用以下8k引擎**
         # • 8k_zh：中文电话通讯；
         # • 8k_en：英文电话通讯；
-        # • 8k_zh_large：中文电话场景专用大模型引擎【大模型版】。通过显著提升模型参数规模与语言建模能力，实现对电话音频中复杂场景（如口音干扰、背景噪声）的高精度识别，识别准确率较常规版本大幅提升。[点击这里](https://console.cloud.tencent.com/asr/demonstrate) 对比常规版本与普方大模型效果；
+        # • 8k_zh_large：中文电话场景专用大模型引擎【大模型版】。当前模型同时支持中文、上海话、四川话、武汉话、贵阳话、昆明话、西安话、郑州话、太原话、兰州话、银川话、西宁话、南京话、合肥话、南昌话、长沙话、苏州话、杭州话、济南话、天津话、石家庄话、黑龙江话、吉林话、辽宁话、闽南语、客家话、粤语、南宁话方言识别，通过显著提升模型参数规模与语言建模能力，实现对电话音频中复杂场景（如口音干扰、背景噪声）的高精度识别，识别准确率较常规版本大幅提升。[点击这里](https://console.cloud.tencent.com/asr/demonstrate) 对比常规版本与普方大模型效果；
 
         # 注意：如您有电话通讯场景识别需求，但发现需求语种仅支持16k，可将8k音频传入下方16k引擎，亦能获取识别结果。但**16k引擎并非基于电话通讯数据训练，无法承诺此种调用方式的识别效果，需由您自行验证识别结果是否可用**
 
         # 通用场景引擎：
         # **注意：除电话通讯场景以外的其它识别场景，请务必使用以下16k引擎**
-        # • **16k_zh_en**：中英粤+7种方言大模型引擎【大模型版】。当前模型同时支持中文、英语、粤语、四川、陕西、河南、上海、湖南、湖北和安徽方言识别，模型参数量极大，语言模型性能增强，针对噪声大、回音大、人声小、人声远等低质量音频的识别准确率极大提升;
+        # • **16k_zh_en**：中英粤+9种方言大模型引擎【大模型版】。当前模型同时支持中文、英语、粤语、四川、陕西、河南、上海、湖南、湖北、安徽、闽南和潮汕方言识别，模型参数量极大，语言模型性能增强，针对噪声大、回音大、人声小、人声远等低质量音频的识别准确率极大提升;
         # • **16k_zh_large：**普方英大模型引擎【大模型版】。当前模型同时支持中文、英文、[多种中文方言](https://cloud.tencent.com/document/product/1093/35682)等语言的识别，模型参数量极大，语言模型性能增强，针对噪声大、回音大、人声小、人声远等低质量音频的识别准确率极大提升，[点击这里](https://console.cloud.tencent.com/asr/demonstrate) 对比中文普通话常规版本与普方英大模型版本的识别效果；
         # • **16k_multi_lang：**多语种大模型引擎【大模型版】。当前模型同时支持英语、日语、韩语、阿拉伯语、菲律宾语、法语、印地语、印尼语、马来语、葡萄牙语、西班牙语、泰语、土耳其语、越南语、德语的识别，可实现15个语种的自动识别(句子/段落级别)；
         # • **16k_zh：**中文普通话通用引擎，支持中文普通话和少量英语，使用丰富的中文普通话语料训练，覆盖场景广泛，适用于除电话通讯外的所有中文普通话识别场景；
@@ -432,7 +432,8 @@ module TencentCloud
         # @type CallbackUrl: String
         # @param SpeakerDiarization: 是否开启说话人分离
         # 0：不开启；
-        # 1：开启（仅支持以下引擎：8k_zh/16k_zh/16k_ms/16k_en/16k_id/16k_zh_large/16k_zh_dialect/16k_zh_en，且ChannelNum=1时可用）；
+        # 1：开启（仅支持以下引擎：8k_zh/8k_zh_large/16k_zh/16k_ms/16k_en/16k_id/16k_zh_large/16k_zh_dialect/16k_zh_en，且ChannelNum=1时可用）；
+        # 3: 开启角色分离，需配合SpeakerRoles参数使用（增值服务，仅支持16k_zh_en引擎，可支持传入声纹对录音文件内的说话人进行角色认证）
         # 默认值为 0
 
         # 注意：
@@ -537,13 +538,21 @@ module TencentCloud
         # 注意：
         # 1. 本功能配置完成后，预计在10分钟后生效
         # @type ReplaceTextId: String
-
-        attr_accessor :EngineModelType, :ChannelNum, :ResTextFormat, :SourceType, :Data, :DataLen, :Url, :CallbackUrl, :SpeakerDiarization, :SpeakerNumber, :HotwordId, :ReinforceHotword, :CustomizationId, :EmotionRecognition, :EmotionalEnergy, :ConvertNumMode, :FilterDirty, :FilterPunc, :FilterModal, :SentenceMaxLength, :Extra, :HotwordList, :KeyWordLibIdList, :ReplaceTextId
+        # @param SpeakerRoles: 开启角色分离能力
+        # 配合SpeakerDiarization: 3 使用，ASR增值服务，可传入一组声纹信息进行角色认证，仅支持16k_zh_en引擎。
+        # 需传入SpeakerRoleInfo数据组，确定说话人的角色信息，涉及RoleAudioUrl和RoleName两个参数。
+        # RoleAudioUrl：需要认证角色的声纹音频地址，建议30s内的纯净人声，最长不能超过45s。
+        # RoleName：需要认证角色的名称，若匹配成功，会替换话者分离中的SpeakerID。
+        # 示例：
+        # "{\"EngineModelType\":\"16k_zh_en\",\"ChannelNum\":1,\"ResTextFormat\":1,\"SourceType\":0,\"Url\":\"需要进行ASR识别的音频链接\",\"SpeakerDiarization\":3,\"SpeakerRoles\":[{\"RoleAudioUrl\":\"需要认证角色的声纹音频地址\",\"RoleName\":\"需要认证角色的名称\"}]}"
+        # @type SpeakerRoles: Array
+
+        attr_accessor :EngineModelType, :ChannelNum, :ResTextFormat, :SourceType, :Data, :DataLen, :Url, :CallbackUrl, :SpeakerDiarization, :SpeakerNumber, :HotwordId, :ReinforceHotword, :CustomizationId, :EmotionRecognition, :EmotionalEnergy, :ConvertNumMode, :FilterDirty, :FilterPunc, :FilterModal, :SentenceMaxLength, :Extra, :HotwordList, :KeyWordLibIdList, :ReplaceTextId, :SpeakerRoles
         extend Gem::Deprecate
-        deprecate :ReinforceHotword, :none, 2025, 11
-        deprecate :ReinforceHotword=, :none, 2025, 11
+        deprecate :ReinforceHotword, :none, 2025, 12
+        deprecate :ReinforceHotword=, :none, 2025, 12
 
-        def initialize(enginemodeltype=nil, channelnum=nil, restextformat=nil, sourcetype=nil, data=nil, datalen=nil, url=nil, callbackurl=nil, speakerdiarization=nil, speakernumber=nil, hotwordid=nil, reinforcehotword=nil, customizationid=nil, emotionrecognition=nil, emotionalenergy=nil, convertnummode=nil, filterdirty=nil, filterpunc=nil, filtermodal=nil, sentencemaxlength=nil, extra=nil, hotwordlist=nil, keywordlibidlist=nil, replacetextid=nil)
+        def initialize(enginemodeltype=nil, channelnum=nil, restextformat=nil, sourcetype=nil, data=nil, datalen=nil, url=nil, callbackurl=nil, speakerdiarization=nil, speakernumber=nil, hotwordid=nil, reinforcehotword=nil, customizationid=nil, emotionrecognition=nil, emotionalenergy=nil, convertnummode=nil, filterdirty=nil, filterpunc=nil, filtermodal=nil, sentencemaxlength=nil, extra=nil, hotwordlist=nil, keywordlibidlist=nil, replacetextid=nil, speakerroles=nil)
           @EngineModelType = enginemodeltype
           @ChannelNum = channelnum
           @ResTextFormat = restextformat
@@ -568,6 +577,7 @@ module TencentCloud
           @HotwordList = hotwordlist
           @KeyWordLibIdList = keywordlibidlist
           @ReplaceTextId = replacetextid
+          @SpeakerRoles = speakerroles
         end
 
         def deserialize(params)
@@ -595,6 +605,14 @@ module TencentCloud
           @HotwordList = params['HotwordList']
           @KeyWordLibIdList = params['KeyWordLibIdList']
           @ReplaceTextId = params['ReplaceTextId']
+          unless params['SpeakerRoles'].nil?
+            @SpeakerRoles = []
+            params['SpeakerRoles'].each do |i|
+              speakerroleinfo_tmp = SpeakerRoleInfo.new
+              speakerroleinfo_tmp.deserialize(i)
+              @SpeakerRoles << speakerroleinfo_tmp
+            end
+          end
         end
       end
 
@@ -931,8 +949,8 @@ module TencentCloud
 
         attr_accessor :TagInfos, :Offset, :Limit
         extend Gem::Deprecate
-        deprecate :TagInfos, :none, 2025, 11
-        deprecate :TagInfos=, :none, 2025, 11
+        deprecate :TagInfos, :none, 2025, 12
+        deprecate :TagInfos=, :none, 2025, 12
 
         def initialize(taginfos=nil, offset=nil, limit=nil)
           @TagInfos = taginfos
@@ -1056,8 +1074,8 @@ module TencentCloud
 
         attr_accessor :TagInfos, :Limit, :Offset
         extend Gem::Deprecate
-        deprecate :TagInfos, :none, 2025, 11
-        deprecate :TagInfos=, :none, 2025, 11
+        deprecate :TagInfos, :none, 2025, 12
+        deprecate :TagInfos=, :none, 2025, 12
 
         def initialize(taginfos=nil, limit=nil, offset=nil)
           @TagInfos = taginfos
@@ -1357,8 +1375,8 @@ module TencentCloud
 
         attr_accessor :ModelName, :DictName, :ModelId, :ModelType, :ServiceType, :ModelState, :AtUpdated, :TagInfos
         extend Gem::Deprecate
-        deprecate :TagInfos, :none, 2025, 11
-        deprecate :TagInfos=, :none, 2025, 11
+        deprecate :TagInfos, :none, 2025, 12
+        deprecate :TagInfos=, :none, 2025, 12
 
         def initialize(modelname=nil, dictname=nil, modelid=nil, modeltype=nil, servicetype=nil, modelstate=nil, atupdated=nil, taginfos=nil)
           @ModelName = modelname
@@ -1638,14 +1656,14 @@ module TencentCloud
 
         attr_accessor :EngSerViceType, :SourceType, :VoiceFormat, :ProjectId, :SubServiceType, :Url, :UsrAudioKey, :Data, :DataLen, :WordInfo, :FilterDirty, :FilterModal, :FilterPunc, :ConvertNumMode, :HotwordId, :CustomizationId, :ReinforceHotword, :HotwordList, :InputSampleRate
         extend Gem::Deprecate
-        deprecate :ProjectId, :none, 2025, 11
-        deprecate :ProjectId=, :none, 2025, 11
-        deprecate :SubServiceType, :none, 2025, 11
-        deprecate :SubServiceType=, :none, 2025, 11
-        deprecate :UsrAudioKey, :none, 2025, 11
-        deprecate :UsrAudioKey=, :none, 2025, 11
-        deprecate :ReinforceHotword, :none, 2025, 11
-        deprecate :ReinforceHotword=, :none, 2025, 11
+        deprecate :ProjectId, :none, 2025, 12
+        deprecate :ProjectId=, :none, 2025, 12
+        deprecate :SubServiceType, :none, 2025, 12
+        deprecate :SubServiceType=, :none, 2025, 12
+        deprecate :UsrAudioKey, :none, 2025, 12
+        deprecate :UsrAudioKey=, :none, 2025, 12
+        deprecate :ReinforceHotword, :none, 2025, 12
+        deprecate :ReinforceHotword=, :none, 2025, 12
 
         def initialize(engservicetype=nil, sourcetype=nil, voiceformat=nil, projectid=nil, subservicetype=nil, url=nil, usraudiokey=nil, data=nil, datalen=nil, wordinfo=nil, filterdirty=nil, filtermodal=nil, filterpunc=nil, convertnummode=nil, hotwordid=nil, customizationid=nil, reinforcehotword=nil, hotwordlist=nil, inputsamplerate=nil)
           @EngSerViceType = engservicetype
@@ -1824,6 +1842,26 @@ module TencentCloud
         end
       end
 
+      # 说话人注册角色声纹信息
+      class SpeakerRoleInfo < TencentCloud::Common::AbstractModel
+        # @param RoleAudioUrl: 音频url地址，建议不超过30秒，最大45秒
+        # @type RoleAudioUrl: String
+        # @param RoleName: 不超过30字节
+        # @type RoleName: String
+
+        attr_accessor :RoleAudioUrl, :RoleName
+
+        def initialize(roleaudiourl=nil, rolename=nil)
+          @RoleAudioUrl = roleaudiourl
+          @RoleName = rolename
+        end
+
+        def deserialize(params)
+          @RoleAudioUrl = params['RoleAudioUrl']
+          @RoleName = params['RoleName']
+        end
+      end
+
       # [录音文件识别](https://cloud.tencent.com/document/product/1093/37823#3.-.E8.BE.93.E5.87.BA.E5.8F.82.E6.95.B0)、[实时语音异步识别](https://cloud.tencent.com/document/product/1093/52061#3.-.E8.BE.93.E5.87.BA.E5.8F.82.E6.95.B0)请求的返回数据
       class Task < TencentCloud::Common::AbstractModel
         # @param TaskId: 任务ID，可通过此ID在轮询接口获取识别状态与结果。TaskId数据类型为**uint64**。

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: tencentcloud-sdk-asr
 version: !ruby/object:Gem::Version
-  version: 3.0.1161
+  version: 3.0.1184
 platform: ruby
 authors:
 - Tencent Cloud
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2025-11-04 00:00:00.000000000 Z
+date: 2025-12-18 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: tencentcloud-sdk-common
@@ -34,8 +34,8 @@ extensions: []
 extra_rdoc_files: []
 files:
 - lib/tencentcloud-sdk-asr.rb
-- lib/v20190614/client.rb
 - lib/v20190614/models.rb
+- lib/v20190614/client.rb
 - lib/VERSION
 homepage: https://github.com/TencentCloud/tencentcloud-sdk-ruby
 licenses: