提出了一种基于Hilbert-Huang变换的语音信号基音周期提取方法。该方法无须对语音信号进行分帧截断。语音信号直接进行Hilbert-Huang变换后,通过基音频率搜索处理得到基音频率及其随时间的变化。实验结果表明,与传统的基音周期提取方法相比,该方法既能真实描述语音信号的非平稳非线性特性,又能提高基音周期提取的准确性和分辨率。 A pitch period extraction method of speech signal based on Hilbert-Huang transform is proposed. The method does not need to frame the voice signal truncation. After the speech signal is directly subjected to Hilbert-Huang transform, the pitch frequency and its variation with time are obtained through the pitch frequency search processing. The experimental results show that compared with the traditional pitch extraction method, the proposed method can not only describe the non-stationary nonlinearity of the speech signal but also improve the accuracy and resolution of the pitch extraction.