首页> 外国专利> Method and system for decomposing an acoustic signal into sound objects, as well as a sound object and its use

Method and system for decomposing an acoustic signal into sound objects, as well as a sound object and its use

机译:用于将声学信号分解为声音对象的方法和系统,以及声音对象及其用途

摘要

The object of the invention is a method and a system for decomposition of acoustic signal into sound objects having the form of signals with slowly-varying amplitude and frequency, as well as sound objects and their use. The system for decomposition of acoustic signal into sound objects having the form of sinusoidal waveforms with slowly varying amplitude and frequency, comprising a sub-system for determining parameters of a short term signal model and a sub-system for determining parameters of a long term signal model based on said parameters, wherein said subsystem for determining short term parameters comprises a converter system for conversion of the analogue acoustic signal into a digital input signal P IN characterized in that said subsystem for determining short term parameters further comprises a filter bank (20) with filter central frequencies distributed according to logarithmic distribution, each digital filter having a window length proportionally to the central frequency wherein each filter (20) is adapted to determine a real value FC (n) and an imaginary value FS (n) of said filtered signal, said filter bank (2) being connected to a system for tracking objects (3), wherein said system for tracking objects (3) comprises a spectrum analysing system (31) adapted to detect all constituent elements of the input signal P IN , a voting system (32) adapted to determine the frequency of all detected constituent elements based on maximum values of the function FG(n) resulting from a mathematical operation reflecting the number of neighbouring filters (20) which output an angular frequency value substantially similar to an angular frequency value of each consecutive filter (20), and in that said subsystem for determining long term parameters comprises a system for associating objects (33), a shape forming system (37) adapted to determine characteristic points describing slowly-varying sinusoidal waveforms, an active objects database (34) and a sound objects database (35).
机译:发明内容本发明的目的是一种用于将声信号分解为具有振幅和频率逐渐变化的信号形式的声音对象的方法和系统,以及声音对象及其用途。用于将声信号分解为振幅和频率呈缓慢变化的正弦波形形式的声音对象的系统,包括用于确定短期信号模型参数的子系统和用于确定长期信号参数的子系统基于所述参数的模型,其中所述用于确定短期参数的子系统包括用于将模拟声信号转换为数字输入信号P IN的转换器系统,其特征在于,所述用于确定短期参数的子系统还包括滤波器组(20)具有根据对数分布分布的滤波器中心频率,每个数字滤波器的窗口长度与中心频率成比例,其中每个滤波器(20)适于确定所述滤波后的实数值FC(n)和虚数值FS(n)信号,所述滤波器组(2)连接到用于跟踪对象(3)的系统,其中所述系统用于r个跟踪对象(3)包括适用于检测输入信号P IN的所有组成元素的频谱分析系统(31),适用于根据函数的最大值确定所有检测到的组成元素的频率的表决系统(32)由数学运算得到的FG(n)反映了相邻滤波器(20)的数量,该相邻滤波器输出与每个连续滤波器(20)的角频率值基本相似的角频率值,并且所述用于确定长期参数的子系统包括用于关联对象的系统(33),适于确定描述缓慢变化的正弦波形的特征点的形状形成系统(37),活动对象数据库(34)和声音对象数据库(35)。

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号