公开(公告)号：US20150220618A1

公开(公告)日：2015-08-06

申请号：US14608737

申请日：2015-01-29

Applicant: Verint Systems Ltd.

Inventor： Yair Horesh ,  Roni Romano ,  Uri Segal ,  Omer Ziv

IPC: G06F17/30 ,  G06N99/00

CPC classification number: G06F17/30598 ,  G06F17/30734 ,  G06N5/022 ,  G06Q30/01 ,  G10L2015/088

Abstract: Systems, methods, and media for developing ontologies and analyzing communication data are provided herein. In an example implementation, the method includes: identifying terms in in a set of communication data; identifying a list of possible relations of the identified terms; scoring the possible relations according to a set of predefined merits; ranking the possible relations into a list of possible relations in descending order according to their score; and tagging relations in the set of communication data. The relations may be tagged by identifying the possible relations in the communication data in order corresponding with the list of possible relations. The possible relations that have lower rankings that conflict with higher ranking relations are not tagged. The conflicts may be determined by a predefined set of conflict criteria.

Abstract translation: 本文提供了用于开发本体和分析通信数据的系统，方法和媒体。 在示例实现中，该方法包括：识别一组通信数据中的术语; 确定所识别术语的可能关系的列表; 根据一组预定义的优点对可能的关系进行评分; 根据他们的得分将可能的关系排列成可能的关系列表，降序排列; 并在该组通信数据中标记关系。 可以通过根据可能的关系列表来确定通信数据中的可能关系来标记关系。 与较高排名关系冲突的排名较低的可能关系未被标记。 冲突可以通过预定义的一组冲突标准来确定。

公开(公告)号：US20140074467A1

公开(公告)日：2014-03-13

申请号：US14016783

申请日：2013-09-03

Applicant: Verint Systems Ltd.

Inventor： Omer Ziv ,  Ron Wein ,  Ido Shapira ,  Ran Achituv

IPC: G10L15/26

CPC classification number: G10L15/26 ,  G10L17/06 ,  G10L25/51 ,  G10L25/78 ,  G10L2025/783

Abstract: The system and method of separating speakers in an audio file including obtaining an audio file. The audio file is transcribed into at least one text file by a transcription server. Homogenous speech segments are identified within the at least one text file. The audio file is segmented into homogenous audio segments that correspond to the identified homogenous speech segments. The homogenous audio segments of the audio file are separated into a first speaker audio file and second speaker audio file the first speaker audio file and the second speaker audio file are transcribed to produce a diarized transcript.

Abstract translation: 分离音频文件中的扬声器的系统和方法，包括获取音频文件。 音频文件由转录服务器转录成至少一个文本文件。 在至少一个文本文件内识别均匀的语音段。 音频文件被分割成与所识别的同源语音片段对应的同质音频段。 音频文件的同质音频片段被分成第一扬声器音频文件和第二扬声器音频文件，第一扬声器音频文件和第二扬声器音频文件被转录以产生经过缩小的转录。