Speech recognition semantic classification training
    1.
    发明授权
    Speech recognition semantic classification training 有权
    语音识别语义分类训练

    公开(公告)号:US08781833B2

    公开(公告)日:2014-07-15

    申请号:US12460249

    申请日:2009-07-15

    IPC分类号: G10L15/00

    摘要: An automated method is described for developing an automated speech input semantic classification system such as a call routing system. A set of semantic classifications is defined for classification of input speech utterances, where each semantic classification represents a specific semantic classification of the speech input. The semantic classification system is trained from training data having little or no in-domain manually transcribed training data, and then operated to assign input speech utterances to the defined semantic classifications. Adaptation training data based on input speech utterances is collected with manually assigned semantic labels. When the adaptation training data satisfies a pre-determined adaptation criteria, the semantic classification system is automatically retrained based on the adaptation training data.

    摘要翻译: 描述了用于开发诸如呼叫路由系统的自动化语音输入语义分类系统的自动化方法。 定义了一组语义分类用于输入语音语音的分类,其中每个语义分类表示语音输入的特定语义分类。 语义分类系统由具有很少或没有域内手动转录训练数据的训练数据训练,然后操作以将输入语音话语分配给定义的语义分类。 基于输入语音语音的适应性训练数据采用手动分配的语义标签进行收集。 当适应训练数据满足预定的适应标准时,语义分类系统将根据适应训练数据自动重新训练。

    Speech recognition semantic classification training
    2.
    发明申请
    Speech recognition semantic classification training 有权
    语音识别语义分类训练

    公开(公告)号:US20100023331A1

    公开(公告)日:2010-01-28

    申请号:US12460249

    申请日:2009-07-15

    摘要: An automated method is described for developing an automated speech input semantic classification system such as a call routing system. A set of semantic classifications is defined for classification of input speech utterances, where each semantic classification represents a specific semantic classification of the speech input. The semantic classification system is trained from training data having little or no in-domain manually transcribed training data, and then operated to assign input speech utterances to the defined semantic classifications. Adaptation training data based on input speech utterances is collected with manually assigned semantic labels. When the adaptation training data satisfies a pre-determined adaptation criteria, the semantic classification system is automatically retrained based on the adaptation training data.

    摘要翻译: 描述了用于开发诸如呼叫路由系统的自动化语音输入语义分类系统的自动化方法。 定义了一组语义分类用于输入语音语音的分类,其中每个语义分类表示语音输入的特定语义分类。 语义分类系统由具有很少或没有域内手动转录训练数据的训练数据训练,然后操作以将输入语音话语分配给定义的语义分类。 基于输入语音语音的适应性训练数据采用手动分配的语义标签进行收集。 当适应训练数据满足预定的适应标准时,语义分类系统将根据适应训练数据自动重新训练。

    Universally tagged frequent call-routing user queries as a knowledge base for reuse across applications
    4.
    发明授权
    Universally tagged frequent call-routing user queries as a knowledge base for reuse across applications 有权
    通用标记频繁的呼叫路由用户查询作为跨应用程序重用的知识库

    公开(公告)号:US08406384B1

    公开(公告)日:2013-03-26

    申请号:US13352640

    申请日:2012-01-18

    IPC分类号: H04M1/64

    摘要: A computer-implemented method is described for developing query tags for classification of user queries to a call routing application. Multiple user query corpuses are accessed that contain user queries from multiple call routing applications in multiple different vertical domains. A set of frequent user queries are selected which appear in multiple different query corpuses in multiple different vertical domains. From these are developed frequent query tags for semantic classification of the frequent user queries. The frequent query tags are then stored in a query tag database.

    摘要翻译: 描述了用于开发用于将呼叫路由应用的用户查询分类的查询标签的计算机实现的方法。 访问多个用户查询语料库,其中包含来自多个不同垂直域中的多个呼叫路由应用程序的用户查询。 选择一组频繁的用户查询,这些查询显示在多个不同的垂直域中的多个不同的查询语料库中。 从这些开发的频繁查询标签用于频繁用户查询的语义分类。 然后,频繁的查询标签存储在查询标签数据库中。