Patents by Inventor Christopher S. Co
Christopher S. Co has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Publication number: 20230386652Abstract: 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 one or more utterances; processing the input acoustic sequence using a speech recognition model to generate a transcription of the input acoustic sequence, wherein the speech recognition model comprises a domain-specific language model; and providing the generated transcription of the input acoustic sequence as input to a domain-specific predictive model to generate structured text content that is derived from the transcription of the input acoustic sequence.Type: ApplicationFiled: August 15, 2023Publication date: November 30, 2023Inventors: Christopher S. Co, Navdeep Jaitly, Lily Hao Yi Peng, Katherine Irene Chou, Ananth Sankar
-
Patent number: 11763936Abstract: 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 one or more utterances; processing the input acoustic sequence using a speech recognition model to generate a transcription of the input acoustic sequence, wherein the speech recognition model comprises a domain-specific language model; and providing the generated transcription of the input acoustic sequence as input to a domain-specific predictive model to generate structured text content that is derived from the transcription of the input acoustic sequence.Type: GrantFiled: December 4, 2020Date of Patent: September 19, 2023Assignee: Google LLCInventors: Christopher S. Co, Navdeep Jaitly, Lily Hao Yi Peng, Katherine Irene Chou, Ananth Sankar
-
Publication number: 20210090724Abstract: 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 one or more utterances; processing the input acoustic sequence using a speech recognition model to generate a transcription of the input acoustic sequence, wherein the speech recognition model comprises a domain-specific language model; and providing the generated transcription of the input acoustic sequence as input to a domain-specific predictive model to generate structured text content that is derived from the transcription of the input acoustic sequence.Type: ApplicationFiled: December 4, 2020Publication date: March 25, 2021Inventors: Christopher S. Co, Navdeep Jaitly, Lily Hao Yi Peng, Katherine Irene Chou, Ananth Sankar
-
Patent number: 10860685Abstract: 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 one or more utterances; processing the input acoustic sequence using a speech recognition model to generate a transcription of the input acoustic sequence, wherein the speech recognition model comprises a domain-specific language model; and providing the generated transcription of the input acoustic sequence as input to a domain-specific predictive model to generate structured text content that is derived from the transcription of the input acoustic sequence.Type: GrantFiled: November 28, 2016Date of Patent: December 8, 2020Assignee: Google LLCInventors: Christopher S. Co, Navdeep Jaitly, Lily Hao Yi Peng, Katherine Irene Chou, Ananth Sankar
-
Publication number: 20180150605Abstract: 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 one or more utterances; processing the input acoustic sequence using a speech recognition model to generate a transcription of the input acoustic sequence, wherein the speech recognition model comprises a domain-specific language model; and providing the generated transcription of the input acoustic sequence as input to a domain-specific predictive model to generate structured text content that is derived from the transcription of the input acoustic sequence.Type: ApplicationFiled: November 28, 2016Publication date: May 31, 2018Inventors: Christopher S. Co, Navdeep Jaitly, Lily Hao Yi Peng, Katherine Irene Chou, Ananth Sankar
-
Patent number: 9916070Abstract: The present invention pertains to geographical image processing of time-dependent imagery. Various assets acquired at different times are stored and processing according to acquisition date in order to generate one or more image tiles for a geographical region of interest. The different image tiles are sorted based on asset acquisition date. Multiple image tiles for the same region of interest may be available. In response to a user request for imagery as of a certain date, one or more image tiles associated with assets from prior to that date are used to generate a time-based geographical image for the user.Type: GrantFiled: September 30, 2014Date of Patent: March 13, 2018Assignee: Google LLCInventors: Reuel William Nash, Christopher S. Co, Andrew Kirmse
-
Patent number: 9245383Abstract: Aspects of the present disclosure relate to building volumetric data structures for intersection testing. For example, 3D data may be represented by points associated with triangles. The triangles may be rasterized to a 3D grid. Each cell of the grid may contain a set of triangles. The grid may be used to generate a new grid of larger grid cells, where each larger grid cell represents some portion of cells of the original grid. The triangle data from each new cell may then be encoded as a list of integers including the triangle data. The list of values may be run-length-encoded. The result is a single octree cube. This process may be repeated in order to generate additional octree cubes for the volumetric structure. The volumetric structure may then be used to identify triangles that intersect with a given ray.Type: GrantFiled: May 20, 2013Date of Patent: January 26, 2016Assignee: Google Inc.Inventor: Christopher S. Co
-
Publication number: 20150154796Abstract: Aspects of the present disclosure relate to building volumetric data structures for intersection testing. For example, 3D data may be represented by points associated with triangles. The triangles may be rasterized to a 3D grid. Each cell of the grid may contain a set of triangles. The grid may be used to generate a new grid of larger grid cells, where each larger grid cell represents some portion of cells of the original grid. The triangle data from each new cell may then be encoded as a list of integers including the triangle data. The list of values may be run-length-encoded. The result is a single octree cube. This process may be repeated in order to generate additional octree cubes for the volumetric structure. The volumetric structure may then be used to identify triangles that intersect with a given ray.Type: ApplicationFiled: May 20, 2013Publication date: June 4, 2015Applicant: GOOGLE INC.Inventor: Christopher S. Co
-
Publication number: 20140365861Abstract: A web page having content and instructions is stored on a tangible non-transitory computer-readable medium. A browser application executes on a processor of a client device. When the browser application interprets the instructions in the web page, the instructions cause the browser application to display the content on the client device, transfer binary data to a system cache of the client device prior to the functionality of a browser plugin being invoked, where the binary data is used only by the browser plugin configured to operate in the browser application, and invoke functionality of the browser plugin, so that the browser plugin accesses the binary data via the system cache during execution.Type: ApplicationFiled: April 25, 2011Publication date: December 11, 2014Applicant: GOOGLE INC.Inventors: Vermont Lasmarias, Christopher S. Co, Mihai Mudure
-
Patent number: 8872847Abstract: The present invention pertains to geographical image processing of time-dependent imagery. Various assets acquired at different times are stored and processing according to acquisition date in order to generate one or more image tiles for a geographical region of interest. The different image tiles are sorted based on asset acquisition date. Multiple image tiles for the same region of interest may be available. In response to a user request for imagery as of a certain date, one or more image tiles associated with assets from prior to that date are used to generate a time-based geographical image for the user.Type: GrantFiled: September 9, 2010Date of Patent: October 28, 2014Assignee: Google Inc.Inventors: Reuel William Nash, Christopher S. Co, Andrew Kirmse
-
Publication number: 20110007094Abstract: The present invention pertains to geographical image processing of time-dependent imagery. Various assets acquired at different times are stored and processing according to acquisition date in order to generate one or more image tiles for a geographical region of interest. The different image tiles are sorted based on asset acquisition date. Multiple image tiles for the same region of interest may be available. In response to a user request for imagery as of a certain date, one or more image tiles associated with assets from prior to that date are used to generate a time-based geographical image for the user.Type: ApplicationFiled: September 9, 2010Publication date: January 13, 2011Applicant: GOOGLE INC.Inventors: Reuel William Nash, Christopher S. Co, Andrew Kirmse