摘要
耳鸣主要表现为非外部声音产生的听觉感知。因其发病率较高、严重影响人的正常生活而越来越受到关注。研究表明,相似而不重复的音乐可作为耳鸣康复音来缓解耳鸣症状。但现有的合成音乐存在无法与试听者个性化匹配和试听效果较差等问题。针对这些问题,基于超混沌算法与主旋律提取,提出一种个性化合成耳鸣康复音的新方法。结果表明,此方法合成的康复音在保证整体相似不重复的前提下,试听效果更舒缓自然,且在耳鸣模型中对耳鸣有一定抑制作用,对耳鸣的治疗具有一定的参考价值。
Tinnitus is mainly manifested as auditory perception without external sounds. It gets more and more attention because of its high incidence and serious impact on people＇s normal life. Scientific researches show that similar but nonerepetitive music can be used as tinnitus rehabilitation to relieve tinnitus symptoms. However,the current synthetic music can＇t personalizedly match with the audio test and audition effect is poor. To solve the problems,this paper proposes a method of personalized synthesizing tinnitus rehabilitation sound based on hyper-chaos and main melody extraction. Results show that the tinnitus rehabilitation sound is not only self-similar but also more soothing and natural.It has a certain inhibitory effect on tinnitus in the tinnitus model.Thus,it has a referential significance for tinnitus treatment.
作者
金宁敏
何培宇
潘帆
陈杰梅
JIN Ning-min;HE Pei-yu;PAN Fan;CHEN Jie-mei(School of Electronic Information and Engineering,Sichuan University,Chengdu 610065,China)
出处
《成都信息工程大学学报》
2018年第4期359-364,共6页
Journal of Chengdu University of Information Technology
基金
四川省科技支撑资助项目(2011SZ0123
2013GZ1043)
关键词
信号与信息处理
语音合成
耳鸣康复音
个性化
超混沌
主旋律提取
signal and information processing
speech synthesis
tinnitus rehabilitation sound
personalized
hyper-chaos
main melody extraction
