论文部分内容阅读
近年来,随着通信事业的发展,特别是移动通信和计算机通信的发展,话音检测技术已作为语言信号处理的一个重要课题,得到了人们普遍的重视,不少科学家在此领域作出了巨大的贡献,提出了一些从背景噪声中识别话音的方法,如音节检测法、电压检测法、基音检测法等。这些方法均能在某种程度上正确识别话音信号的有无,但是可靠性还不高,漏检率或误检率较高,前端剪音也较严重,不能很好满足系统对话音检测的要求。为了提高话音检
In recent years, with the development of communication, especially the development of mobile communication and computer communication, speech detection technology has been regarded as an important issue in speech signal processing and has gained universal attention. Many scientists have made huge Contributions, some proposed from the background noise in the voice recognition methods, such as syllable detection method, voltage detection method, pitch detection method. All of these methods can correctly identify the presence of voice signals to a certain extent, but the reliability is not high, the rate of missed or false detection is high, and the front-end clipping is more severe, which can not meet the requirements of the system for voice detection Claim. In order to improve the voice check