论文部分内容阅读
采用话音激活检测(VoicedActivityDetection,VAD)技术的目的是检测语音通信时是否有话音存在,检测到静音时加以抑制,使其不占用或极少占用信道带宽,检测到话音时才对其进行压缩编码与传输。鲁棒性语音识别系统、数字移动通信和因特网实时语音传输等领域要求在恶劣声学环境条件下进行VAD检测,以节省带宽并抑制噪声,因此VAD技术是目前语音处理领域的重要问题。文中给出的几种最新VAD算法(EZCR-VAD,STAT-VAD和E-VAD)是在低信噪比环境下的话音检测具有很好的鲁棒性的算法。
The purpose of VoicedActivityDetection (VAD) is to detect whether voice is present during voice communication, to suppress silence when it is detected, so that it does not occupy or occupy a minimum of channel bandwidth, and only encode voice when it is detected With transmission. Robust speech recognition systems, digital mobile communications and real-time voice over the Internet require VAD detection in harsh acoustic environments to save bandwidth and suppress noise, so VAD technology is an important issue in today’s speech processing area. The latest VAD algorithms (EZCR-VAD, STAT-VAD and E-VAD) presented in this paper are algorithms that have good robustness to speech detection in low SNR environments.