Speech emotion recognition based on parametric filter and fractal dimension

In this paper, we propose a new method that employs two novel features, correlation density (C_d) and fractal dimension (F_d), to recognize emotional states contained in speech. The former feature obtained by a list of parametric filters reflects the broad frequency components and the fine structure of lower frequency components, contributed by unvoiced phones and voiced phones, respectively; the latter feature indicates the nonlinearity and self-similarity of a speech signal. Comparative experiments based on Hidden Markov Model and K Nearest Neighbor methods are carried out. The results show that C_d and F_d are much more closely related with emotional expression than the features commonly used.