发明名称 CAPTION SYSTEM BASED ON UTTERANCE RECOGNITION
摘要 PROBLEM TO BE SOLVED: To provide a method for adding a closed caption to a television program that uses caption data as a broadcasting source, and is not currently embedded to a broadcasting transmission stream. SOLUTION: The method for displaying text information, corresponding to an utterance portion in the voice signal of the television program as a closed caption on a video display device, includes a process for decoding the voice signal in the video program, a process for filtering the voice signal for extracting the utterance portion, a process for analyzing the utterance portion to each utterance component according to an utterance model for grouping the analyzed utterance component, a process for identifying a word in database corresponding to the grouped utterance component, and a process for converting the identified word to the text data to display the identified word on the display device as the closed caption.
申请公布号 JP2002300495(A) 申请公布日期 2002.10.11
申请号 JP20010352435 申请日期 2001.11.16
申请人 MATSUSHITA ELECTRIC IND CO LTD 发明人 KERN MICHAEL
分类号 H04N5/278;G10L15/00;G10L15/02;G10L15/14;G10L15/26;H04N5/445;H04N5/60;(IPC1-7):H04N5/60 主分类号 H04N5/278
代理机构 代理人
主权项
地址