类 SpeechConfig

定义语音/意向识别或语音合成配置的类。

成员

EmbeddedSpeechConfig

语法: public friend EmbeddedSpeechConfig;

HybridSpeechConfig

语法: public friend HybridSpeechConfig;

运算符 SPXSPEECHCONFIGHANDLE

语法: public inline explicit operator SPXSPEECHCONFIGHANDLE ( ) const;

用于获取基础句柄值的内部运算符。

返回

句柄。

SetSpeechRecognitionLanguage

语法: public inline void SetSpeechRecognitionLanguage ( const std::string & lang );

将输入语言设置为语音识别器。

参数

  • lang 指定要以 BCP-47 格式识别的口语的名称。

GetSpeechRecognitionLanguage

语法: public inline std::string GetSpeechRecognitionLanguage ( ) const;

获取语音识别的输入语言。 语言以 BCP-47 格式指定。

返回

语音识别语言。

SetSpeechSynthesisLanguage

语法: public inline void SetSpeechSynthesisLanguage ( const std::string & lang );

设置语音合成器的语言。 在版本 1.4.0 中添加。

参数

  • lang 指定语言 (的名称,例如 en-US)

GetSpeechSynthesisLanguage

语法: public inline std::string GetSpeechSynthesisLanguage ( ) const;

获取语音合成器的语言。 在版本 1.4.0 中添加。

返回

语音合成语言。

SetSpeechSynthesisVoiceName

语法: public inline void SetSpeechSynthesisVoiceName ( const std::string & voiceName );

设置语音合成器的声音。 在版本 1.4.0 中添加。

参数

  • voiceName 指定语音的名称

GetSpeechSynthesisVoiceName

语法: public inline std::string GetSpeechSynthesisVoiceName ( ) const;

获取语音合成器的声音。 在版本 1.4.0 中添加。

返回

语音合成语音名称。

SetSpeechSynthesisOutputFormat

语法: public inline void SetSpeechSynthesisOutputFormat ( SpeechSynthesisOutputFormat formatId );

设置语音合成输出格式 (例如 Riff16Khz16BitMonoPcm) 。 在版本 1.4.0 中添加。

参数

  • formatId 指定输出格式 ID

GetSpeechSynthesisOutputFormat

语法: public inline std::string GetSpeechSynthesisOutputFormat ( ) const;

获取语音合成输出格式。 在版本 1.4.0 中添加。

返回

语音合成输出格式。

SetEndpointId

语法: public inline void SetEndpointId ( const std::string & endpointId );

设置自定义语音或自定义语音的终结点 ID。

参数

  • endpointId 终结点 ID。

GetEndpointId

语法: public inline std::string GetEndpointId ( ) const;

获取自定义语音或自定义语音的终结点 ID。

返回

终结点 ID。

SetAuthorizationToken

语法: public inline void SetAuthorizationToken ( const std::string & token );

设置要连接到服务的授权令牌。 注意:调用方需要确保授权令牌有效。 在授权令牌过期之前,调用方需要使用新的有效令牌调用此资源库来刷新它。 由于在创建新的识别器时会复制配置值,因此新令牌值将不适用于已创建的识别器。 对于以前创建的识别器,需要设置相应识别器的授权令牌才能刷新令牌。 否则,识别器将在识别过程中遇到错误。

参数

  • token 授权令牌。

GetAuthorizationToken

语法: public inline std::string GetAuthorizationToken ( ) const;

获取用于连接到服务的授权令牌。

返回

授权令牌。

GetSubscriptionKey

语法: public inline std::string GetSubscriptionKey ( ) const;

获取用于创建语音识别器、意向识别器、翻译识别器或语音合成器的订阅密钥。

返回

订阅密钥。

GetRegion

语法: public inline std::string GetRegion ( ) const;

获取用于创建语音识别器、意向识别器、翻译识别器或语音合成器的区域键。

返回

区域。

GetOutputFormat

语法: public inline OutputFormat GetOutputFormat ( ) const;

获取语音识别输出格式 (简单或详细) 。 注意:此输出格式适用于语音识别结果,请使用 SpeechConfig::GetSpeechSynthesisOutputFormat 获取合成的音频输出格式。

返回

语音识别输出格式。

SetOutputFormat

语法: public inline void SetOutputFormat ( OutputFormat format );

设置语音识别输出格式 (简单或详细) 。 注意:此输出格式适用于语音识别结果,请使用 SpeechConfig::SetSpeechSynthesisOutputFormat 设置合成的音频输出格式。

参数

  • format 语音识别输出格式

SetProfanity

语法: public inline void SetProfanity ( ProfanityOption profanity );

设置不雅内容选项。 在版本 1.5.0 中添加。

参数

  • profanity 不雅选项值。

EnableAudioLogging

语法: public inline void EnableAudioLogging ( );

在服务中启用音频日志记录。 在版本 1.5.0 中添加。

音频和内容日志存储在 Microsoft 拥有的存储中,或存储在与认知服务订阅关联的自己的存储帐户中, (自带存储 (BYOS) 已启用语音资源) 。

RequestWordLevelTimestamps

语法: public inline void RequestWordLevelTimestamps ( );

在响应结果中包含单词级时间戳。 在版本 1.5.0 中添加。

EnableDictation

语法: public inline void EnableDictation ( );

启用听写模式。 仅在语音连续识别中受支持。 在版本 1.5.0 中添加。

SetProxy

语法: public inline void SetProxy ( const std::string & proxyHostName , uint32_t proxyPort , const std::string & proxyUserName , const std::string & proxyPassword );

设置在版本 1.1.0 中添加的代理配置。

注意:代理功能在 macOS 上不可用。 此函数对此平台没有影响。

参数

  • proxyHostName 代理服务器的主机名,不使用协议方案 ([http://](http://))

  • proxyPort 代理服务器的端口号

  • proxyUserName 代理服务器的用户名

  • proxyPassword 代理服务器的密码

SetProperty

语法: public inline void SetProperty ( const std::string & name , const std::string & value );

按名称设置属性值。

参数

  • name 属性名称。

  • value 属性值。

GetProperty

语法: public inline std::string GetProperty ( const std::string & name ) const;

按名称获取属性值。

参数

  • name 参数名称。

返回

属性值。

GetProperty

语法: public inline std::string GetProperty ( PropertyId id ) const;

按 ID 获取属性值。

参数

  • id 参数 ID。

返回

属性值。

SetProperty

语法: public inline void SetProperty ( PropertyId id , const std::string & value );

按 ID 设置属性值。

参数

  • id 属性 ID。

  • value 属性值。

SetServiceProperty

语法: public inline void SetServiceProperty ( const std::string & name , const std::string & value , ServicePropertyChannel channel );

设置将使用指定通道传递给服务的属性值。 在版本 1.5.0 中添加。

参数

  • name 属性名称。

  • value 属性值。

  • channel 用于将指定属性传递给服务的通道。

~SpeechConfig

语法: public inline virtual ~SpeechConfig ( );

解构对象。

FromSubscription

语法: public inline static std::shared_ptr< SpeechConfig > FromSubscription ( const std::string & subscription , const std::string & region );

使用指定的订阅密钥和区域创建语音配置的实例。

参数

  • subscription 订阅密钥。

  • region 区域名称 (请参阅 区域页) 。

返回

指向新语音配置实例的共享指针。

FromAuthorizationToken

语法: public inline static std::shared_ptr< SpeechConfig > FromAuthorizationToken ( const std::string & authToken , const std::string & region );

使用指定的授权令牌和区域创建语音配置的实例。 注意:调用方需要确保授权令牌有效。 在授权令牌过期之前,调用方需要使用新的有效令牌调用此资源库来刷新它。 由于在创建新的识别器时会复制配置值,因此新令牌值将不适用于已创建的识别器。 对于以前创建的识别器,需要设置相应识别器的授权令牌才能刷新令牌。 否则,识别器将在识别过程中遇到错误。

参数

  • authToken 授权令牌。

  • region 区域名称 (请参阅 区域页) 。

返回

指向新语音配置实例的共享指针。

FromEndpoint

语法: public inline static std::shared_ptr< SpeechConfig > FromEndpoint ( const std::string & endpoint , const std::string & subscription );

使用指定的终结点和订阅创建语音配置的实例。 此方法仅适用于使用非标准服务终结点的用户。 注意:终结点 URI 中指定的查询参数不会更改,即使它们是由任何其他 API 设置的。 例如,如果在 URI 中将识别语言定义为查询参数“language=de-DE”,并且还由 SetSpeechRecognitionLanguage (“en-US”) 设置,则 URI 中的语言设置优先,有效语言为“de-DE”。 其他 API 只能设置终结点 URI 中未指定的参数。 注意:若要将授权令牌与 FromEndpoint 配合使用,请使用 FromEndpoint (const SPXSTRING&) ,然后在创建的 SpeechConfig 实例上调用 SetAuthorizationToken () 。

参数

  • endpoint 要连接到的服务终结点。

  • subscription 订阅密钥。

返回

指向新语音配置实例的共享指针。

FromEndpoint

语法: public inline static std::shared_ptr< SpeechConfig > FromEndpoint ( const std::string & endpoint );

使用指定的终结点创建 SpeechConfig 的实例。 此方法仅适用于使用非标准服务终结点的用户。 注意:终结点 URI 中指定的查询参数不会更改,即使它们是由任何其他 API 设置的。 特定查询参数是否受支持取决于终结点和方案。 例如,如果在 URI 中将识别语言定义为查询参数“language=de-DE”,并且还由 SetSpeechRecognitionLanguage (“en-US”) 设置,则 URI 中的语言设置优先,有效语言为“de-DE”。 仅当终结点和方案组合支持语言作为查询参数时,该示例才适用。 其他 API 只能设置终结点 URI 中未指定的参数。 注意:如果终结点需要订阅密钥进行身份验证,请使用 FromEndpoint (const SPXSTRING&、const SPXSTRING&) 将订阅密钥作为参数传递。 若要将授权令牌与 FromEndpoint 配合使用,请使用此方法创建 SpeechConfig 实例,然后在创建的 SpeechConfig 实例上调用 SetAuthorizationToken () 。 注意:在版本 1.5.0 中添加。

参数

  • endpoint 要连接到的服务终结点 URI。

返回

指向新语音配置实例的共享指针。

FromHost

语法: public inline static std::shared_ptr< SpeechConfig > FromHost ( const std::string & host , const std::string & subscription );

使用指定的主机和订阅创建语音配置的实例。 此方法仅适用于使用非默认服务主机的用户。 将假定使用标准资源路径。 对于具有非标准资源路径或根本没有路径的服务,请改用 FromEndpoint。 注意:主机 URI 中不允许使用查询参数,必须由其他 API 设置。 注意:若要对 FromHost 使用授权令牌,请使用 FromHost (const SPXSTRING&) ,然后在创建的 SpeechConfig 实例上调用 SetAuthorizationToken () 。 注意:在版本 1.8.0 中添加。

参数

  • host 要连接到的服务主机。 格式为“protocol://host:port”,其中“:p ort”是可选的。

  • subscription 订阅密钥。

返回

指向新语音配置实例的共享指针。

FromHost

语法: public inline static std::shared_ptr< SpeechConfig > FromHost ( const std::string & host );

使用指定的主机创建 SpeechConfig 的实例。 此方法仅适用于使用非默认服务主机的用户。 将假定使用标准资源路径。 对于具有非标准资源路径或根本没有路径的服务,请改用 FromEndpoint。 注意:主机 URI 中不允许使用查询参数,必须由其他 API 设置。 注意:如果主机需要订阅密钥进行身份验证,请使用 FromHost (const SPXSTRING&、const SPXSTRING&) 将订阅密钥作为参数传递。 若要将授权令牌与 FromHost 配合使用,请使用此方法创建 SpeechConfig 实例,然后在创建的 SpeechConfig 实例上调用 SetAuthorizationToken () 。 注意:在版本 1.8.0 中添加。

参数

  • host 要连接到的服务主机 URI。 格式为“protocol://host:port”,其中“:p ort”是可选的。

返回

指向新语音配置实例的共享指针。