公开(公告)号：US10566009B1

公开(公告)日：2020-02-18

申请号：US16520633

申请日：2019-07-24

Applicant: Google LLC

Inventor： Sourish Chaudhuri ,  Achal D. Dave ,  Bryan Andrew Seybold

IPC: G10L25/57 ,  G06F16/638 ,  G10L17/00 ,  G10L15/06 ,  G10L17/04 ,  G10L17/26 ,  G10L15/04 ,  G10L15/01 ,  G06K9/00

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for audio classifiers. In one aspect, a method includes obtaining a plurality of video frames from a plurality of videos, wherein each of the plurality of video frames is associated with one or more image labels of a plurality of image labels determined based on image recognition; obtaining a plurality of audio segments corresponding to the plurality of video frames, wherein each audio segment has a specified duration relative to the corresponding video frame; and generating an audio classifier trained using the plurality of audio segment and the associated image labels as input, wherein the audio classifier is trained such that the one or more groups of audio segments are determined to be associated with respective one or more audio labels.

公开(公告)号：US10381022B1

公开(公告)日：2019-08-13

申请号：US15041379

申请日：2016-02-11

Applicant: Google LLC

Inventor： Sourish Chaudhuri ,  Achal D. Dave ,  Bryan Andrew Seybold

IPC: G06K9/00 ,  G10L15/01 ,  G10L15/04 ,  G10L15/06 ,  G10L17/00 ,  G10L17/04 ,  G10L17/26 ,  G10L25/57 ,  G06F16/638

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for audio classifiers. In one aspect, a method includes obtaining a plurality of video frames from a plurality of videos, wherein each of the plurality of video frames is associated with one or more image labels of a plurality of image labels determined based on image recognition; obtaining a plurality of audio segments corresponding to the plurality of video frames, wherein each audio segment has a specified duration relative to the corresponding video frame; and generating an audio classifier trained using the plurality of audio segment and the associated image labels as input, wherein the audio classifier is trained such that the one or more groups of audio segments are determined to be associated with respective one or more audio labels.