Speech feature processing method, device, electronic equipment and storage medium

Abstract

Embodiments of the present application provide a speech feature processing method, device, electronic equipment, and storage medium, wherein the method includes: removing noise speech features from noisy speech features to obtain pure speech feature estimates; It is estimated to perform coding processing to obtain first speech feature coding information, and to perform coding processing on the speech features of the noisy speech to obtain second speech feature coding information; according to the first speech feature coding information and the second speech feature encoding information to obtain target speech feature encoding information for decoding. The embodiments of the present application can accurately form speech feature coding information for noisy speech, and provide a basis for improving the accuracy of speech recognition.

Description

technical field [0001] The embodiments of the present application relate to the technical field of voice recognition, and in particular to a voice feature processing method, device, electronic equipment, and storage medium. Background technique [0002] Speech recognition is a technology that converts speech into text. It is widely used in human-computer voice interaction, intelligent control, communication and other scenarios. Therefore, it is of great significance to improve the accuracy of speech recognition. [0003] When performing speech recognition, it is necessary to encode the speech features corresponding to the speech to be recognized to form deep speech feature coding information, and then perform further processing such as decoding the speech feature coding information to realize text conversion; however, in the actual application scene of speech recognition In addition to the pure speech of the speaker itself, the speech to be recognized may also have noise, th...

AI Technical Summary

Problems solved by technology

[0003] When performing speech recognition, it is necessary to encode the speech features corresponding to the speech to be recognized to form deep speech feature coding information, and then perform further processing such as decoding the speech feature coding information to realize text conversion; however, in the actual application scene of speech recognition In addition to the pure speech of the speaker itself, the speech to be recognized may also have noise, that is, the speech to be recognized is a noisy speech, which makes the speech features extracted from the noisy speech have noise speech features, resulting in the speech formed by subsequent processing The feature coding information also has correspondingly noisy coding information, which greatly affects the accuracy of speech recognition and reduces the user experience of speech recognition services;

Images