公开(公告)号：US20130179161A1

公开(公告)日：2013-07-11

申请号：US13781055

申请日：2013-02-28

Applicant: Broadcom Corporation

Inventor： Robert W. Zopf ,  Kelly Hale

IPC: G10L21/00

CPC classification number: G10L21/00 ,  G10L17/00 ,  G10L19/0018 ,  G10L19/16

Abstract: A communications network is used to transfer user attribute information about participants in a communication session to their respective communication terminals for storage and use thereon to configure a speech codec to operate in a speaker-dependent manner, thereby improving speech coding efficiency. In a network-assisted model, the user attribute information is stored on the communications network and selectively transmitted to the communication terminals while in a peer-assisted model, the user attribute information is derived by and transferred between communication terminals.

Abstract translation: 通信网络用于将通信会话中的参与者的用户属性信息传送到其各自的通信终端以进行存储和使用，以将语音编解码器配置为以说话者相关的方式操作，从而提高语音编码效率。 在网络辅助模型中，用户属性信息存储在通信网络上，并且在对等辅助模型中选择性地发送到通信终端，用户属性信息由通信终端之间导出和传送。

公开(公告)号：US08781080B2

公开(公告)日：2014-07-15

申请号：US13959636

申请日：2013-08-05

Applicant: Broadcom Corporation

Inventor： Kelly Hale ,  Jes Thyssen

IPC: H04M11/00 ,  G10L15/26 ,  H04L12/66

CPC classification number: G10L15/26 ,  H04M3/42221 ,  H04M3/53333 ,  H04M11/10 ,  H04M2201/40 ,  H04M2201/60

Abstract: Systems and methods for presenting audio messages are provided. In some aspects, a method includes receiving an audio message from a first user and generating a text-based representation of the audio message. The method also includes generating one or more identification tags based on the text-based representation of the audio message. At least one of the one or more identification tags includes a subject of the audio message. The method also includes presenting at least one of the text-based representation of the audio message or the one or more identification tags to a second user using a graphical user interface.

Abstract translation: 提供用于呈现音频消息的系统和方法。 在一些方面，一种方法包括从第一用户接收音频消息并生成音频消息的基于文本的表示。 该方法还包括基于音频消息的基于文本的表示来生成一个或多个识别标签。 一个或多个识别标签中的至少一个包括音频消息的对象。 该方法还包括使用图形用户界面向第二用户呈现音频消息或一个或多个识别标签的基于文本的表示中的至少一个。