Self-learning spectrogram feature extraction method for speech emotion recognition

Abstract

The invention discloses a self-learning spectrogram feature extraction method for speech emotion recognition. The method is characterized by, to begin with, carrying out preprocessing on speech, of which the emotion is known, in a standard corpus to obtain a quantitative spectrogram gray level image; then, calculating a Gabor spectrogram of the obtained spectrogram gray level image; carrying out training on an extracted LBP statistical histogram by utilizing a recognizable characteristic learning algorithm and constructing a global significance pattern set with different scales and different directions; and finally, carrying out feature selection on the LBP statistical histograms of the Gabor spectrograms under different scales and different directions of the speech by utilizing the global significance pattern set to obtain processed statistical histograms, and cascading the N statistical histograms to obtain speech emotion characteristics suitable for emotion classification. The emotion features can recognize different types of emotions, and recognition rate thereof is substantially superior to that of existing acoustic features.

Description

technical field [0001] The invention relates to the technical field of speech emotion recognition, in particular to a feature extraction method applied to a speech emotion recognition system. Background technique [0002] Voice, as one of the most important means of communication, has received more and more attention in the new field of human-computer interaction. In order to make the human-computer interaction system and the dialogue system of the robot more intelligent and perfect, the emotional analysis of voice is becoming more and more important. important. In addition, in some long-term, monotonous, high-intensity tasks (such as spaceflight, navigation, etc.), relevant personnel often have certain negative emotions. Effective identification of these negative emotions helps to improve individual cognition and work Efficiency, prevent problems before they happen. Speech emotion recognition can also play an important role in criminal investigation and intelligent assist...

AI Technical Summary

Problems solved by technology

However, the research on the correlation of speech signals often only studies one domain in the frequency domain or time domain, and there are few literatures that combine the time-frequency domain correlation of speech signals for research.

Images