Tone recognition method and apparatus, electronic device and computer-readable storage medium

Abstract

The invention discloses a tone recognition method and apparatus, an electronic device and a computer-readable storage medium. The method includes the steps of acquiring a sample text; acquiring a plurality of recording data for the sample text, the plurality of recording data being speeches obtained by reading the sample text by using a plurality of types of tone; extracting tone features in the plurality of recording data, and determining tone types corresponding to the tone features; and establishing a tone recognition model based on the sample text, the tone features in the plurality of recording data, and the corresponding tone types. The method can make each sample text be marked with the features of various types of tone and corresponding tone types so that in online application, themethod can perform speech recognition on user-entered speech data and also the tone recognition of speech data through the pre-established tone recognition model at the same time, thereby realizing the recognition of various types of tone in the speech data and improving the speech recognition effect.

Description

technical field [0001] The present invention relates to the technical field of speech recognition, in particular to a tone recognition method, device, electronic equipment and computer-readable storage medium. Background technique [0002] At present, some voice texts are the same, but because the user uses different tones when speaking the voice text, the user's intention is greatly different. The text can be spoken in a rhetorical tone, and since the user uses two different tones, the user's intent will also be different. However, the traditional voice recognition technology can only recognize the content of the user's voice, but cannot recognize the interrogative tone of the user's voice, which makes the voice recognition effect poor, resulting in poor dialogue between man-machine voice. Contents of the invention [0003] The object of the present invention is to solve one of the above-mentioned technical problems at least to a certain extent. [0004] For this reason...

AI Technical Summary

Problems solved by technology

[0002] At present, some voice texts are the same, but because the user uses different tones when speaking the voice text, the user's intention is greatly different. The text can be spoken in a rhetorical tone, and since the user uses two different tones, the user's intent will also be different

However, the traditional voice recognition technology can only recognize the content of the user's voice, but cannot recognize the interrogative tone of the user's voice, which makes the voice recognition effect poor, resulting in poor dialogue between man-machine voice

Images