Character extraction method, text reading method, dialogue text generation method, apparatus and device, and storage medium

Abstract

The invention relates to a character extraction method, a text reading method, a dialogue type text generation method and device, equipment and a storage medium, the character extraction method comprises the following steps: receiving a target text, the target text comprising one or more sentences; inputting the target text into a pre-trained language model, and determining a speaker corresponding to an utterance in the target text by using the language model; wherein the step of determining the speaker corresponding to the utterance in the target text by utilizing the language model comprises the following steps of: inferring the speaker corresponding to the utterance in the target text by utilizing the context of a sentence in a text input into the language model; wherein the process of training the language model comprises the step of inferring a speaker corresponding to an utterance in a training text by utilizing a context of a sentence in the training text input into the language model. By means of the character extraction method, the speaker corresponding to the utterance in the text can be accurately extracted and recognized.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a character extraction method, a text reading method, a dialogue text generation method, a device, a device and a storage medium. Background technique [0002] The statements herein merely provide background information related to the present disclosure and do not necessarily constitute prior art. [0003] Existing character recognition and extraction methods for text content can only extract the characters appearing in the text, but cannot confirm who said the words appearing in the text. It is impossible to accurately extract the attributable character of the sentence with the spoken content from the text, which cannot be achieved by traditional methods. Contents of the invention [0004] The object of the present invention is to provide a new character extraction method, device, equipment and storage medium, so as to accurately extract and identify the speaker cor...

AI Technical Summary

Problems solved by technology

[0003] Existing character recognition and extraction methods for text content can only extract the characters appearing in the text, but cannot confirm who said the words appearing in the text

Accurately extract the attributable character of the sentence with the spoken content from the text, which cannot be achieved by traditional methods

Images