发明名称 SPEECH DETECTOR, SPEECH DETECTION PROGRAM AND SPEECH DETECTION METHOD
摘要 PROBLEM TO BE SOLVED: To provide a speech detector capable of accurately detecting a speech from a moving image.SOLUTION: The speech detector includes: an image acquisition section; an extraction section; a calculation section; and a determination section. The image acquisition section acquires a moving image of a face as a target of speech detection. The extraction section extracts a mouth area in the acquired moving image. The calculation section calculates an aspect ratio of the extracted mouth area. The determination section determines whether a movement of the mouth represents a speech or a non-speech based on variation in changing frequency in the calculated aspect ratio.SELECTED DRAWING: Figure 1
申请公布号 JP2016091324(A) 申请公布日期 2016.05.23
申请号 JP20140225441 申请日期 2014.11.05
申请人 FUJITSU LTD 发明人 IMAI TAKESHI
分类号 G06T7/20;G06T1/00;G06T7/00;G10L15/04 主分类号 G06T7/20
代理机构 代理人
主权项
地址