基于HHT瞬时能频值的含噪耳语音声韵分割

Initial/Final Segmentation Based on Instantaneous Energy Frequency Value of HHT for Noisy Whispered Speech

提出了一种基于希尔伯特-黄变换瞬时能频值的含噪耳语音声韵分割算法。根据耳语音声韵母幅频特性,运用HHT,分离出耳语音中高频部分的瞬时幅值与频率,同步去除低频噪声,提取出能够反映声韵母过渡信息的特征参数——瞬时能频值,利用该参数对耳语音进行声韵分割。实验结果表明:与相对熵算法相比,该算法对含噪耳语音进行的声韵分割正确率较高,能够较准确地进行耳语音声韵分割。

An algorithm of noisy whispered speech initial/final segmentation is proposed,which is based on instantaneous energy-frequency value（IEFV） of Hilbert-Huang Transform（HHT）.With analysis of the amplitude-frequency characteristics of Chinese whispers,HHT is applied to separating instantaneous amplitude and instantaneous frequency from whispers while reducing the noise of low-frequency domain,and extracting IEFV.For IEFV could show the difference between initial and final whisper effectively,it is used as the feature for initial/final segmentation.The experiments indicate that the correct initial,final segmentation rate of noisy whisper by this algorithm is higher than by the symmetric relative entropy function algorithm,and this proposed algorithm is feasible and efficient.