Speech-emotion recognition method based on improved fuzzy vector quantization

Abstract

The invention discloses a speech-emotion recognition method based on improved fuzzy vector quantization. The method extends the sum of fuzzy membership function from one to N so as to reduce the influence of sample wild-point on an iteration-training process to a certain extent, and adopts a clustering method based on similarity threshold and a minimum distance principle in the iteration-training process so as to avoid the problem that a clustering center is sensitive to initial values and easy to fall into local minimum values to a certain extent. Experimental results prove that the method can effectively improve the emotion recognition rate of the prior fuzzy vector quantization method.

Description

technical field [0001] The invention relates to a speech recognition method, in particular to a speech emotion recognition system and method. Background technique [0002] The speech emotion automatic recognition technology mainly includes two problems: one is to use the features in the speech signal as emotion recognition, that is, the problem of emotional feature extraction, including feature extraction and selection; the other is how to classify specific speech data, That is, the problem of pattern recognition, including various pattern recognition algorithms, such as nearest neighbors, neural networks, support vector machines, etc. [0003] The emotional features used in speech emotion recognition are mainly prosody parameters and sound quality parameters. The former includes duration, speech rate, energy, pitch frequency and its derivative parameters, and the latter is mainly formant, harmonic-to-noise ratio and its derivative parameters. wait. According to the theory...

AI Technical Summary

Problems solved by technology

Before the present invention, among the various existing identification methods, although the neural network method has a high degree of nonlinearity and strong classification ability, the required learning time increases rapidly with the increase of the network, and the local minimum The problem is also a shortcoming; the hidden Markov method (HMM) takes a long time to establish and train, and it needs to solve the problem of high computational complexity when it is applied in practice

Although the quadratic discriminant algorithm is simple and has a small amount of calculation, it must be based on the premise that the feature vector obeys the normal distribution, which greatly affects the recognition rate.

The recognition method based on vector quantization is rarely used due to problems such as quantization error and initial value sensitivity. Although fuzzy vector quantization alleviates the problem of quantization error to a certain extent, it is still easy to fall into the problem of initial value sensitivity and local minimum.

Images