公开(公告)号：US20250061887A1

公开(公告)日：2025-02-20

申请号：US18938222

申请日：2024-11-05

Applicant: GOOGLE LLC

Inventor： Mark Bowers ,  Brian F. Allen ,  Nida Zada ,  Julie Anne Seguin

IPC: G10L13/04 ,  G06V40/10 ,  G10L13/10

Abstract: Generating synthesized speech audio data on behalf of a given user in a conversation. The synthesized speech audio data includes synthesized speech that incorporates textual segment(s). The textual segment(s) can include recognized text that results from processing spoken input, of the given user, using a speech recognition model and/or can include a selection of a rendered suggestion that conveys the textual segment(s). Some implementations dynamically determine one or more prosodic properties for use in speech synthesis of the textual segment, and generate the synthesized speech with the one or more determined prosodic properties. The prosodic properties can be determined based on the textual segment(s) used in speech synthesis, textual segment(s) corresponding to recent spoken input of additional participant(s), attribute(s) of relationship(s) between the given user and additional participant(s) in the conversation, and/or feature(s) of a current location for the conversation.

公开(公告)号：US12190859B2

公开(公告)日：2025-01-07

申请号：US17792012

申请日：2020-02-10

Applicant: GOOGLE LLC

Inventor： Mark Bowers ,  Brian F. Allen ,  Nida Zada ,  Julie Anne Seguin

IPC: G10L13/10 ,  G10L13/04 ,  G06V40/10

Abstract: Generating synthesized speech audio data on behalf of a given user in a conversation. The synthesized speech audio data includes synthesized speech that incorporates textual segment(s). The textual segment(s) can include recognized text that results from processing spoken input, of the given user, using a speech recognition model and/or can include a selection of a rendered suggestion that conveys the textual segment(s). Some implementations dynamically determine one or more prosodic properties for use in speech synthesis of the textual segment, and generate the synthesized speech with the one or more determined prosodic properties. The prosodic properties can be determined based on the textual segment(s) used in speech synthesis, textual segment(s) corresponding to recent spoken input of additional participant(s), attribute(s) of relationship(s) between the given user and additional participant(s) in the conversation, and/or feature(s) of a current location for the conversation.

公开(公告)号：US20190130192A1

公开(公告)日：2019-05-02

申请号：US15798733

申请日：2017-10-31

Applicant: Google LLC

Inventor： Alejandro Kauffmann ,  Andrew Dahley ,  Phuong Le ,  Mark Bowers ,  Ignacio Garcia Dorado ,  Robin Debreuil ,  William Lindmeier ,  Brian Allen ,  Ashley Ma ,  Pascal Getreuer

IPC: G06K9/00 ,  H04N5/14 ,  G11B27/036 ,  H04N5/262

CPC classification number: G06K9/00751 ,  G06K9/00765 ,  G11B27/031 ,  G11B27/036 ,  G11B27/28 ,  G11B27/34 ,  H04N5/144 ,  H04N5/147 ,  H04N5/2628

Abstract: The present disclosure provides systems and methods that generate a summary storyboard from a plurality of image frames. An example computer-implemented method can include inputting a plurality of image frames into a machine-learned model and receiving as an output of the machine-learned model, object data that describes the respective locations of a plurality of objects recognized in the plurality of image frames. The method can include generating a plurality of image crops that respectively include the plurality of objects and arranging two or more of the plurality of image crops to generate a storyboard.

公开(公告)号：US20230046658A1

公开(公告)日：2023-02-16

申请号：US17792012

申请日：2020-02-10

Applicant: GOOGLE LLC

Inventor： Mark Bowers ,  Brian F. Allen ,  Nida Zada ,  Julie Anne Seguin

IPC: G10L13/04 ,  G10L13/10

Abstract: Generating synthesized speech audio data on behalf of a given user in a conversation. The synthesized speech audio data includes synthesized speech that incorporates textual segment(s). The textual segment(s) can include recognized text that results from processing spoken input, of the given user, using a speech recognition model and/or can include a selection of a rendered suggestion that conveys the textual segment(s). Some implementations dynamically determine one or more prosodic properties for use in speech synthesis of the textual segment, and generate the synthesized speech with the one or more determined prosodic properties. The prosodic properties can be determined based on the textual segment(s) used in speech synthesis, textual segment(s) corresponding to recent spoken input of additional participant(s), attribute(s) of relationship(s) between the given user and additional participant(s) in the conversation, and/or feature(s) of a current location for the conversation.