Children speech emotion recognition method

Abstract

The present invention discloses a children speech emotion recognition method, comprising: creating a training voice database, classifier training and recognizing emotion. The classifier training comprises the step of extracting the speech emotion features, which includes performing pretreatment on the analog speech signal, including sampling, denoising and other basic signal treatments, performing HHT transformation on the pretreated emotion sentences, obtaining the Hilbert spectrum of the emotion speech; performing amplitude feature extraction, intonation feature extraction, tone feature extraction and rhythm feature extraction, and obtaining the emotion feature; reducing dimension of the emotion feature, and obtaining different quaternionic feature of emotion space; performing calculation on the sample set of the training speech database, obtaining the projection of a series of quaternionic features in national emotion state in the emotion space i, thus forming an emotion feature projection set; and carrying out classification training based on the SVM / HMM mixed system by using the data in the emotion feature projection set.

Description

technical field [0001] The invention relates to speech recognition, in particular to a speech emotion recognition method based on children's speech. Background technique [0002] At present, the smart electronic products for children on the market can play songs and sentences, and can also carry out simple conversations through the answering mode, but the sounds made by such electronic products and learning machines are very mechanical, monotonous and unnatural, making them only It can be used as a cold tool, and its humanization function is poor, so it cannot be fully accepted by children. The response mode of existing electronic products cannot make corresponding emotional and style adjustments according to children's emotional changes and children with different personalities. Can play pre-programmed answers with mechanical voices, so the human-computer interaction effect is not good. [0003] In the current research on speech emotion recognition, although there are meth...

AI Technical Summary

Problems solved by technology

On the one hand, the method based on the assumption of framing increases the complexity of the algorithm and slows down the recognition speed. On the other hand, due to the differences between each individual's speech rate and pronunciation characteristics, framing based on time windows often destroys the relationship between speech and semantics. The overall coordination relationship is not conducive to speech recognition, especially not conducive to emotion recognition with a high degree of semantic correlation

[0006] At the same time, in the prior art, there is no research on the unique voice characteristics and emotional characteristics of children, and there is no method for emotional analysis of children's speech and emotional speech synthesis.

Images