Keyword identification method and device, storage medium and computer equipment

Abstract

The invention discloses a keyword identification method and device, a storage medium and computer equipment, and the method comprises the steps: obtaining an original voice signal, and carrying out the framing processing of the original voice signal to obtain a target voice signal, wherein the target voice signal comprises a plurality of voice frames which are arranged according to a time sequence, and the voice frame is represented based on a time domain feature and a frequency domain feature; and inputting the target voice signal into a keyword identification network for keyword identification to obtain a keyword contained in the target voice signal, wherein the keyword identification network comprises a plurality of convolutional network units connected in series and a classifier, the output feature of the previous convolutional network unit is the input feature of the next convolutional network unit, each convolutional network unit comprises one or more convolutional layers, and the classifier is used for classifying the output features of the last convolutional network unit to obtain keywords contained in the target voice signal. The identification precision can be considered while the model parameter quantity is reduced.

Description

Technical field [0001] The present invention relates to the field of artificial intelligence, and more particularly to a keyword identification method and device, a storage medium, a computer device. Background technique [0002] With the development of artificial intelligence, more and more manual intelligence algorithms have been applied to keyword recognition. Keyword Spotting, Briefn KWS is currently an important human-computer interaction method, which has been applied to different electronic products (such as computer, mobile phones, etc.), while some electronic products are smaller than memory capacity, hoping keyword identification The method is covered in less model parameters to reduce the memory used by the algorithm. However, reducing the model parameters used by the keyword identification method may result in a decrease in recognition accuracy. That is, the current keyword identification method cannot balance the accuracy of identification and the amount of model par...

AI Technical Summary

Problems solved by technology

However, reducing the model parameters used by keyword recognition methods may lead to lower recognition accuracy

That is to say, the current keyword recognition method cannot balance the recognition accuracy and the amount of model parameters.

[0003] Compared with traditional algorithms, neural networks can improve the accuracy of keyword recognition and the amount of parameters of the model, such as using depthwise convolution (Depthwise convolution) network for keyword recognition, to Reduce the amount of parameters of the model. However, although the deep convolutional network can reduce the amount of calculation to a certain extent, its number of channels is large, so the overall parameter amount of the network is still large, and it is still not applicable to electronic products with strict memory constraints. And the recognition accuracy of the deep convolutional network still needs to be further improved

Images