基于G.729的自适应实时语音活动检测方法研究

Adaptive and real-time voice activity detection method on G.729

提出了一种运用短时能量和短时过零率两个参数进行静音检测的模型,针对不同情况下的语音数据动态调整模型,实现了不同噪声环境下对语音片段的有效检测。该算法实现简单,可实时的对活动语音进行检测不引入延迟。实验结果表明所采用的方法能够比较准确地检测出语音片断,对于噪声环境和音量低的语音都有很好的自适应性。

On the basis of short time energy and short time cross zero rate,a simple voice activity detection model on G.729 is presented.Some adjust methods are discussed for the model to adapt to different environments automatically.These rules can be realized simply and be used to detect activated voice without any delay.The experimental results show that these rules are effective in voice activity detection and in different voice environments they show better adaptability.