公开(公告)号：US11252114B2

公开(公告)日：2022-02-15

申请号：US15844082

申请日：2017-12-15

Applicant: Google LLC

Inventor： Max Benjamin Braun ,  Nirmal Jitendra Patel

IPC: H04L12/58 ,  G06N20/00

Abstract: A method may involve, for each of one or more messages that are selected from a plurality of messages from an account: (a) extracting one or more phrases from a respective selected message; (b) determining that a conversation includes the respective selected message and one or more other messages from the plurality of messages; (c) generating a first feature vector based on the conversation, wherein the first feature vector includes one or more first features, wherein the one or more first features include one or more words from the conversation; and (d) generating, by a computing system, one or more training-data sets, wherein each training-data set comprises one of the phrases and the first feature vector. The method may further involve: training, by the computing system, a machine-learning application with at least a portion of the one or more training-data sets that are generated for the one or more selected messages; applying the trained machine-learning application to process an incoming message to the account; and responsive to applying the trained machine-learning application, determining one or more reply messages corresponding to the incoming message, wherein the one or more reply messages include at least one of the extracted one or more phrases.

公开(公告)号：US20220131819A1

公开(公告)日：2022-04-28

申请号：US17570927

申请日：2022-01-07

Applicant: Google LLC

Inventor： Max Benjamin Braun ,  Nirmal Jitendra Patel

IPC: H04L51/046 ,  G06N20/00 ,  H04L51/02

Abstract: A method may involve, for each of one or more messages that are selected from a plurality of messages from an account: (a) extracting one or more phrases from a respective selected message; (b) determining that a conversation includes the respective selected message and one or more other messages from the plurality of messages; (c) generating a first feature vector based on the conversation, wherein the first feature vector includes one or more first features, wherein the one or more first features include one or more words from the conversation; and (d) generating, by a computing system, one or more training-data sets, wherein each training-data set comprises one of the phrases and the first feature vector. The method may further involve: training, by the computing system, a machine-learning application with at least a portion of the one or more training-data sets that are generated for the one or more selected messages; applying the trained machine-learning application to process an incoming message to the account; and responsive to applying the trained machine-learning application, determining one or more reply messages corresponding to the incoming message, wherein the one or more reply messages include at least one of the extracted one or more phrases.

公开(公告)号：US10114466B2

公开(公告)日：2018-10-30

申请号：US15238135

申请日：2016-08-16

Applicant: Google LLC

Inventor： Nirmal Patel ,  Hayes Solos Raffle ,  Mat Balez ,  Max Benjamin Braun ,  Jerrica Jones

IPC: G06F3/00 ,  G06F3/01 ,  G06F3/0482 ,  G02B27/01 ,  G06F3/0346

Abstract: Methods and systems for hands-free browsing in a wearable computing device are provided. A wearable computing device may provide for display a view of a first card of a plurality of cards which include respective virtual displays of content. The wearable computing device may determine a first rotation of the wearable computing device about a first axis and one or more eye gestures. Based on a combination of the first rotation and the eye gestures, the wearable computing device may provide for display the navigable menu, which may include an alternate view of the first card and at least a portion of one or more cards. Then, based on a determined second rotation of the wearable computing device about a second axis and based on a direction of the second rotation, the wearable computing device may generate a display indicative of navigation through the navigable menu.

公开(公告)号：US20180109476A1

公开(公告)日：2018-04-19

申请号：US15844082

申请日：2017-12-15

Applicant: Google LLC

Inventor： Max Benjamin Braun ,  Nirmal Jitendra Patel

IPC: H04L12/58 ,  G06N99/00

CPC classification number: H04L51/046 ,  G06N20/00 ,  H04L51/02

Abstract: A method may involve, for each of one or more messages that are selected from a plurality of messages from an account: (a) extracting one or more phrases from a respective selected message; (b) determining that a conversation includes the respective selected message and one or more other messages from the plurality of messages; (c) generating a first feature vector based on the conversation, wherein the first feature vector includes one or more first features, wherein the one or more first features include one or more words from the conversation; and (d) generating, by a computing system, one or more training-data sets, wherein each training-data set comprises one of the phrases and the first feature vector. The method may further involve: training, by the computing system, a machine-learning application with at least a portion of the one or more training-data sets that are generated for the one or more selected messages; applying the trained machine-learning application to process an incoming message to the account; and responsive to applying the trained machine-learning application, determining one or more reply messages corresponding to the incoming message, wherein the one or more reply messages include at least one of the extracted one or more phrases.

公开(公告)号：US20240414112A1

公开(公告)日：2024-12-12

申请号：US18737595

申请日：2024-06-07

Applicant: Google LLC

Inventor： Max Benjamin Braun ,  Nirmal Jitendra Patel

IPC: H04L51/046 ,  G06N20/00 ,  H04L51/02

Abstract: A method may involve, for each of one or more messages that are selected from a plurality of messages from an account: (a) extracting one or more phrases from a respective selected message; (b) determining that a conversation includes the respective selected message and one or more other messages from the plurality of messages; (c) generating a first feature vector based on the conversation, wherein the first feature vector includes one or more first features, wherein the one or more first features include one or more words from the conversation; and (d) generating, by a computing system, one or more training-data sets, wherein each training-data set comprises one of the phrases and the first feature vector. The method may further involve: training, by the computing system, a machine-learning application with at least a portion of the one or more training-data sets that are generated for the one or more selected messages; applying the trained machine-learning application to process an incoming message to the account; and responsive to applying the trained machine-learning application, determining one or more reply messages corresponding to the incoming message, wherein the one or more reply messages include at least one of the extracted one or more phrases.

公开(公告)号：US10083468B2

公开(公告)日：2018-09-25

申请号：US14976466

申请日：2015-12-21

Applicant: Google LLC

Inventor： Steven John Lee ,  Bradley James Rhodes ,  Indika Charles Mendis ,  Max Benjamin Braun

IPC: G06Q30/06 ,  H04L29/06 ,  H04L12/18 ,  H04L12/58 ,  G06F3/16 ,  G06F17/30 ,  G06Q30/04 ,  G06T19/00 ,  G09G5/00

CPC classification number: G06Q30/06 ,  G06F3/167 ,  G06F16/23 ,  G06F16/955 ,  G06Q30/04 ,  G06Q30/0605 ,  G06Q30/0625 ,  G06Q30/0635 ,  G06Q30/0641 ,  G06Q30/0643 ,  G06T19/006 ,  G09G5/006 ,  H04L12/1822 ,  H04L29/06476 ,  H04L51/046 ,  H04L51/12 ,  H04L51/14 ,  H04L65/1069 ,  H04L65/4076

Abstract: Disclosed herein are methods and corresponding systems for facilitating a live registry event between a shopper and one or more remote viewers. In the live registry event, the shopper, registers for gifts that they would like to receive while sharing their registering experience with the viewers in real-time. When the user registers for an item, a server system allows the viewers to purchase the gift for the shopper.

公开(公告)号：US12010079B2

公开(公告)日：2024-06-11

申请号：US17570927

申请日：2022-01-07

Applicant: Google LLC

Inventor： Max Benjamin Braun ,  Nirmal Jitendra Patel

IPC: H04L51/046 ,  G06N20/00 ,  H04L51/02

CPC classification number: H04L51/046 ,  G06N20/00 ,  H04L51/02

Abstract: A method may involve, for each of one or more messages that are selected from a plurality of messages from an account: (a) extracting one or more phrases from a respective selected message; (b) determining that a conversation includes the respective selected message and one or more other messages from the plurality of messages; (c) generating a first feature vector based on the conversation, wherein the first feature vector includes one or more first features, wherein the one or more first features include one or more words from the conversation; and (d) generating, by a computing system, one or more training-data sets, wherein each training-data set comprises one of the phrases and the first feature vector. The method may further involve: training, by the computing system, a machine-learning application with at least a portion of the one or more training-data sets that are generated for the one or more selected messages; applying the trained machine-learning application to process an incoming message to the account; and responsive to applying the trained machine-learning application, determining one or more reply messages corresponding to the incoming message, wherein the one or more reply messages include at least one of the extracted one or more phrases.

公开(公告)号：US20190331914A1

公开(公告)日：2019-10-31

申请号：US15885506

申请日：2018-01-31

Applicant: Google LLC

Inventor： Steven John Lee ,  Indika Charles Mendis ,  Max Benjamin Braun ,  Bradley James Rhodes ,  Casey Ho ,  Michael Patrick Johnson

IPC: G02B27/01 ,  G06F3/16 ,  G06F3/00 ,  G06F3/01

Abstract: An experience sharing session can be established with a wearable computing device. A field of view of an environment can be provided through a head-mounted display (HMD) of the wearable computing device. The HMD is operable to display a computer-generated image overlaying at least a portion of the view. At least one image of the environment can be captured using a camera associated with the wearable computing device. The wearable computing device can receive an indication of a region of interest within the environment via the experience sharing session. The wearable computing device can display, on the HMD, the indication of the region of interest.

公开(公告)号：US09967487B2

公开(公告)日：2018-05-08

申请号：US14952538

申请日：2015-11-25

Applicant: Google LLC

Inventor： Max Benjamin Braun ,  Peter Malkin ,  Sergey Brin

IPC: H04N5/222 ,  H04N5/369 ,  G02B27/01 ,  G06F3/01 ,  G06F3/16 ,  G06T19/00

CPC classification number: H04N5/3698 ,  G02B27/017 ,  G02B2027/0138 ,  G02B2027/014 ,  G02B2027/0178 ,  G06F3/013 ,  G06F3/167 ,  G06T19/006

Abstract: Embodiments may be implemented by a computing device, such as a head-mountable display or mobile phone, in order to pre-emptively warm up the device's camera, when it is probable that a user will be taking a photo. An illustrative method involves a computing device (a) receiving sensor data from one or more sensors associated with the computing device, wherein the computing device comprises an image-capture device, (b) analyzing the sensor data to detect at least one pre-image-capture signal, wherein the at least one pre-image-capture signal indicates a subsequent image-capture signal is likely to be received, and (c) in response to detecting the at least one pre-image-capture signal, causing the computing device to initiate an image-capture preparation process that prepares the image-capture device to capture an image.