摘要
目前在高速发展的电影市场下,偷漏瞒报票房、影片盗录盗放等乱象屡屡出现,严重扰乱了电影市场秩序,因此加强影厅放映内容的监管需求也更为迫切.本文在此背景上将声纹识别技术在电影放映领域进行探索实验,通过在影片播放过程中采集声音样本,并根据Landmark算法构造声纹特征进行匹配,最终达到识别放映内容的结果,为影院放映监测机制提供有效思路.本实验已基本实现该功能并具有较高识别率,但在后期仍需在算法优化改进中做进一步探索.
At present,under the rapid development of the film market,issues such as box office concealment and film piracy have repeatedly appeared,which has seriously disrupted the order of the film market.Therefore,it is urgent to strengthen the supervision of cinema screening content.In this context,this article experiments with voiceprint recognition technology in film screening.Collecting sound samples during film screening,and matching voiceprint features according to the Landmark algorithm,in order to identify the screening content.This method may provide solution for cinema screening monitoring.This experiment has basically achieved the expected function and has a highly recognition rate,the optimization and improvement of the algorithm will be explored in the next stage.
作者
王惠
Wang Hui(China Research Institute of Film Science&Technology)
出处
《现代电影技术》
2020年第3期17-23,共7页
Advanced Motion Picture Technology
基金
国家重点研发计划(National Key R&D Program of China)资助,项目名称《影视媒体融合服务技术集成与应用》,项目编号:2018YFB1404000。
关键词
声纹识别
LANDMARK
电影技术
特征参数提取
Voiceprint recognition
Landmark
Film technology
Characteristic parameter extraction
