Speech recognition method based on neural network stacking autoencoder multi-feature fusion

Abstract

The invention relates to a speech recognition method based on neural network stacking autoencoder multi-feature fusion. Firstly, the original sound data is framed and windowed, and the typical time-domain linear predictive cepstrum coefficient feature and the frequency-domain Mel frequency cepstrum coefficient feature are respectively extracted from the framed and windowed data; then the extractedfeatures are spliced, the initial feature representation vector of acoustic signals is constructed and the training feature library is created; then the multi-layer neural network stacking autoencoder is used for feature fusion and learning; the multi-layer autoencoder adopts the over-limit learning machine algorithm to learn training; and finally the extracted features are trained using the over-limit learning machine algorithm to get the classifier model; and the constructed model is finally used to test sample classification and identification. The method adopts multi-feature fusion basedon the over-limit learning machine multi-layer neural network stacking autoencoder, which has higher recognition accuracy compared with the traditional single feature extraction method.

Description

technical field [0001] The invention relates to the technical field of voice recognition, in particular to a voice recognition method based on neural network stacked autoencoder multi-feature fusion. Background technique [0002] Sound recognition is one of the goals of artificial intelligence, and the ability to accurately identify and classify sound signals plays a key role in the development of artificial intelligence. In the existing technology, the traditional method of feature extraction plus classifier has been widely used. However, the feature extraction should be selected according to different types of sounds. Since the source of sound signals is different, the method of feature extraction is also different. It is necessary to select a feature extraction method with corresponding professional knowledge for different sounds. In addition, the sound situation in some environments is more complicated, and the traditional feature extraction methods cannot achieve the d...

AI Technical Summary

Problems solved by technology

However, this method cannot guarantee the irrelevance between each feature, that is, there is redundancy in the feature vector obtained by direct splicing, which makes the classification effect of the trained model poor.

The low efficiency of model training is a difficulty restricting the field of voice recognition technology

Images