New feature extraction methods in the time-frequency plane for Chinese tone analysis

Since the Chinese tone is highly related to the variation of instantaneous frequency, it is effective to use time-frequency analysis to identify the Chinese tone. In this manuscript, we find that, in addition to the change of instantaneous frequency, the variation of power, the normalized powers in the first and the second harmonics, the frequency turn, and the information of the previous tone are also useful for distinguishing the tone. We also propose a way to identify the 5^th tone. In sum, we adopt 18 features and apply the technique of the support vector machine. Simulations show that the proposed algorithm can accurately identify the Chinese tone.