公开(公告)号：WO2014197303A1

公开(公告)日：2014-12-11

申请号：PCT/US2014/040168

申请日：2014-05-30

Applicant: MICROSOFT CORPORATION

Inventor： CHANG, Shuangyu ,  LEVIT, Michael

IPC: G10L15/32 ,  G10L15/19

CPC classification number: G10L15/065 ,  G10L15/02 ,  G10L15/063 ,  G10L15/19 ,  G10L15/32

Abstract: A received utterance is recognized using different language models. For example, recognition of the utterance is independently performed using a baseline language model (BLM) and using an adapted language model (ALM). A determination is made as to what results from the different language model are more likely to be accurate. Different features may be used to assist in making the determination (e.g. language model scores, recognition confidences, acoustic model scores, quality measurements, ) may be used. A classifier may be trained and then used in determining whether to select the results using the BLM or to select the results using the ALM. A language model may be automatically trained or re-trained that adjusts a weight of the training data used in training the model in response to differences between the two results obtained from applying the different language models.

Abstract translation: 使用不同的语言模型识别接收的话语。 例如，使用基准语言模型（BLM）和使用适应语言模型（ALM）来独立地执行语音的识别。 确定不同语言模型的结果更有可能是准确的。 可以使用不同的特征来辅助确定（例如，语言模型分数，识别信心，声学模型评分，质量测量）。 可以对分类器进行训练，然后用于确定是使用BLM选择结果还是使用ALM选择结果。 可以自动训练或重新训练语言模型，以响应于从应用不同语言模型获得的两个结果之间的差异来调整用于训练模型的训练数据的权重。

公开(公告)号：WO2014190220A2

公开(公告)日：2014-11-27

申请号：PCT/US2014/039258

申请日：2014-05-23

Applicant: MICROSOFT CORPORATION

Inventor： LEVIT, Michael ,  HAKKANI-TUR, Dilek ,  TUR, Gokhan

IPC: G10L15/06

CPC classification number: G10L15/063 ,  G06F17/2818 ,  G06F17/30023 ,  G06F17/30976 ,  G10L15/197

Abstract: A Statistical Machine Translation (SMT) model is trained using pairs of sentences that include content obtained from one or more content sources (e.g. feed(s)) with corresponding queries that have been used to access the content. A query click graph may be used to assist in determining candidate pairs for the SMT training data. All/portion of the candidate pairs may be used to train the SMT model. After training the SMT model using the SMT training data, the SMT model is applied to content to determine predicted queries that may be used to search for the content. The predicted queries are used to train a language model, such as a query language model. The query language model may be interpolated other language models, such as a background language model, as well as a feed language model trained using the content used in determining the predicted queries.

Abstract translation: 使用成对的句子来训练统计机器翻译（SMT）模型，所述句子包括从一个或多个内容源（例如，馈送）获得的内容与已经用于访问内容的对应查询。 可以使用查询点击图来帮助确定SMT训练数据的候选对。 候选对的全部/部分可用于训练SMT模型。 在使用SMT培训数据对SMT模型进行培训后，将SMT模型应用于内容，以确定可能用于搜索内容的预测查询。 预测的查询用于训练语言模型，如查询语言模型。 查询语言模型可以内插其他语言模型，例如背景语言模型，以及使用在确定预测查询中使用的内容训练的馈送语言模型。

公开(公告)号：WO2010141513A2

公开(公告)日：2010-12-09

申请号：PCT/US2010/036964

申请日：2010-06-01

Applicant: MICROSOFT CORPORATION

Inventor： CHANG, Shuangyu ,  LEVIT, Michael ,  BUNTSCHUH, Bruce

IPC: G06F17/00 ,  G06F17/18 ,  G06F17/27 ,  G06F15/16 ,  G06F13/14

CPC classification number: G06N99/005 ,  G10L15/08 ,  G10L15/19 ,  G10L15/32 ,  G10L2015/228

Abstract: Architecture that employs an overall grammar as a set of context-specific grammars for recognition of an input, each responsible for a specific context, such as subtask category, geographic region, etc. The grammars together cover the entire domain. Moreover, multiple recognitions can be run in parallel against the same input, where each recognition uses one or more of the context-specific grammars. The multiple intermediate recognition results from the different recognizer-grammars are reconciled by running re-recognition using a dynamically composed grammar based on the multiple recognition results and potentially other domain knowledge, or selecting the winner using a statistical classifier operating on classification features extracted from the multiple recognition results and other domain knowledge.

Abstract translation: 使用整体语法作为一组上下文特定语法的体系结构，用于识别输入，每个语言负责特定上下文，例如子任务类别，地理区域等。语法一起涵盖整个域。 此外，可以针对相同的输入并行运行多个识别，其中每个识别使用一个或多个上下文特定语法。 通过使用基于多个识别结果和潜在的其他领域知识的动态组合语法运行重新识别来协调来自不同识别器语法的多个中间识别结果，或者使用对从分类特征提取的分类特征进行运算的统计分类器来选择胜者 多重识别结果等领域知识。