公开(公告)号：US09990918B1

公开(公告)日：2018-06-05

申请号：US15788300

申请日：2017-10-19

Applicant: Google Inc.

Inventor： William Chan ,  Navdeep Jaitly ,  Quoc V. Le ,  Oriol Vinyals ,  Noam M. Shazeer

IPC: G10L15/16 ,  G06F17/22 ,  G10L15/183 ,  G10L15/26

CPC classification number: G10L15/16

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media for speech recognition. One method includes obtaining an input acoustic sequence, the input acoustic sequence representing an utterance, and the input acoustic sequence comprising a respective acoustic feature representation at each of a first number of time steps; processing the input acoustic sequence using a first neural network to convert the input acoustic sequence into an alternative representation for the input acoustic sequence; processing the alternative representation for the input acoustic sequence using an attention-based Recurrent Neural Network (RNN) to generate, for each position in an output sequence order, a set of substring scores that includes a respective substring score for each substring in a set of substrings; and generating a sequence of substrings that represent a transcription of the utterance.

公开(公告)号：US09799327B1

公开(公告)日：2017-10-24

申请号：US15055476

申请日：2016-02-26

Applicant: Google Inc.

Inventor： William Chan ,  Navdeep Jaitly ,  Quoc V. Le ,  Oriol Vinyals ,  Noam M. Shazeer

IPC: G10L15/16 ,  G10L15/26 ,  G06F17/22 ,  G10L15/183

CPC classification number: G10L15/16

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media for speech recognition. One method includes obtaining an input acoustic sequence, the input acoustic sequence representing an utterance, and the input acoustic sequence comprising a respective acoustic feature representation at each of a first number of time steps; processing the input acoustic sequence using a first neural network to convert the input acoustic sequence into an alternative representation for the input acoustic sequence; processing the alternative representation for the input acoustic sequence using an attention-based Recurrent Neural Network (RNN) to generate, for each position in an output sequence order, a set of substring scores that includes a respective substring score for each substring in a set of substrings; and generating a sequence of substrings that represent a transcription of the utterance.

公开(公告)号：US09602620B1

公开(公告)日：2017-03-21

申请号：US14278656

申请日：2014-05-15

Applicant: Google Inc.

Inventor： James Roskind ,  Michael Belshe ,  William Chan

IPC: G06F15/16 ,  H04L29/08 ,  G06F17/30

CPC classification number: H04L67/2847 ,  G06F17/30902 ,  G06F2216/13

Abstract: Methods and systems for reducing web page load time include obtaining speculative information associated with a uniform resource locator (URL). The method and system also include determining whether to prefetch content of the URL based on the speculative information. The method and system further include providing an instruction to prefetch the URL content when a determination to prefetch results. The method and system also include prerendering the URL content when the instruction to prefetch the URL content is provided.

公开(公告)号：US08762490B1

公开(公告)日：2014-06-24

申请号：US13681115

申请日：2012-11-19

Applicant: Google Inc.

Inventor： James Roskind ,  Michael Belshe ,  William Chan

IPC: G06F15/16

CPC classification number: H04L67/2847 ,  G06F17/30902 ,  G06F2216/13

Abstract: Methods and systems for reducing web page load time include obtaining speculative information associated with a uniform resource locator (URL). The method and system also include determining whether to prefetch content of the URL based on the speculative information. The method and system further include providing an instruction to prefetch the URL content when a determination to prefetch results. The method and system also include prerendering the URL content when the instruction to prefetch the URL content is provided.

Abstract translation: 减少网页加载时间的方法和系统包括获得与统一资源定位符（URL）相关联的推测信息。 该方法和系统还包括基于投机信息确定是否预取URL的内容。 所述方法和系统还包括当预取结果的确定时提供预取URL内容的指令。 该方法和系统还包括在提供预取URL内容的指令时预先绘制URL内容。