学术
摘 要:
传统的音频时域修正采用SOLA方法,但不适宜处理音频信号中广泛存在的瞬时冲击,存在一定局限性,基于此方法提出了一种改进的SOLA算法,有效地改进了传统算法;或者采用基于正弦模型的时域修正方法,灵活地将不同类型的信号进行不同处理,较好地保留了瞬时冲击特性;仿真实验的结果证明,两种方法均能解决传统SOLA算法中瞬时冲击处理不当的问题,同时两种方法各有利弊。[著者文摘]
文章出处:
《计算机工程与设计》-2008年29卷3期 -680-682页
栏目信息:
分 类 号:
文献标识码:
A
文章编号:
1000-7024（2008）03-0680-03
Research in audio time scale modification
HUANG Mei, LI Shuang-tian （1. Institute of Acoustics, Chinese Academy of Sciences, Beijing 100080, China; 2. Graduate University, Chinese Academy of Sciences, Beijing 100080, China）
Abstract:
The conventional audio time scale modification adopts SOLA （synchronized overlap and add） method, but it has its limitations in that it does not suitable for transients which widely exist in audio signal. Based on this conventional method, a new SOLA algorithm is introduced, which greatly improves conventional method; or adopt time scale method based on sinusoidal model, which flexibly uses different time scale methods according to different type of signals, hence maintains the property oftransients well. The simulation experimental results indicate： both of these two methods can solves the problem of inappropriate transient procession in SOLA method, which respectively have their own advantages.[著者文摘]
Key words:
SOLA; transient; sinusoidal; audio; signal model; time scale
更多>>免费文章
cqvip.com