基于高频混合片段匹配的火焰声音合成  

High Frequency Blend Clips Matching Based Fire Sound Synthesis

作　　者：尹强 刘世光[1,2] Yin Qiang;Liu Shiguang(School of Computer Science and Technology, Tianjin University, Tianjin 300350, China;Tianjin Key Laboratory of Cognitive Computing and Application, Tianjin 300350, China)

机构地区：[1]天津大学计算机科学与技术学院,天津300350 [2]天津市认知计算与应用重点实验室,天津300350

出　　处：《系统仿真学报》2018年第6期2066-2075,共10页Journal of System Simulation

基　　金：国家自然科学基金(61672375;61170118);天津市自然科学基金(14JCQNJC00100)

摘　　要：针对虚拟现实领域火焰声音模拟的问题,提出了一种新的火焰声音合成方法。考虑低频声音物理模型的影响,通过基于物理的可视化火焰模型数据合成火焰的低频信号。对于火焰的高频声音细节,采用基于高频混合片段匹配的方法,按照火焰高频信号的特性生成特定的不同频段的高频噪声信号,并分成相同大小的窗口片段,然后与低频信号进行匹配融合,得到最终的声音信号。通过一些不同物质燃烧的例子以及对比试验验证了新方法的有效性。Aimed at the problem which generated by fire sound simulation in virtual reality field, a new fire sound synthesis method was proposed. Most of the sound synthesis methods currently separate the fire sound into low and high frequencies for synthesizing. The influence of low frequency fire sound models was considered and the low frequency signal was synthesized based on the physical visual fire model data. The high frequency sound signal was synthesized based on high frequency blend clips matching method. According to the characteristics of the fire＇s high frequency, the different frequency bands of high noise signal were generated and separated into the same size windows. The sub-signals with the low frequency signal were matched with the same window size. Various examples and comparisons are presented to verify the plausible method.

关 键 词：火焰 声音 低频 高频 混合片段 

分 类 号：TP391.72[自动化与计算机技术—计算机应用技术]