Method and device for converting text

Abstract

The invention provides a method and device for converting text, including obtaining the target oral text. The target oral text comprises non verbal morpheme and verbal morpheme. The verbal morpheme categories comprises inserted morpheme, repeated morpheme, modified morpheme and wrong beginning morpheme. The characteristics of the target oral text can be determined, which comprises the language probability of words or phrases, lexical feature, part of speech information, semantic information, grammatical information and the conditional probability of adjacent words. The verbal morpheme category of the target oral text is determined according to the text feature of the target oral text. The written language text corresponding to the target oral text is determined according to the verbal morpheme category of the target oral text. The written language text is a text which verbal morpheme of the target oral text is eliminated. The method and device for converting text has the advantage of improving the accuracy of the conversion from oral text to written language text.

Description

technical field [0001] The embodiments of the present invention relate to the field of information technology, and more specifically, to a method and device for converting text. Background technique [0002] Spoken language, as an oral language used in informal occasions (for example, daily conversations, informal speeches, informal document writing, etc.), has the characteristics of informality and flexibility. Written language, as a language used in formal occasions (for example, formal speeches, formal document writing, etc.), has the characteristics of standardization and conciseness. [0003] Because spoken language does not have the characteristics of brevity and standardization that written language has, so in some occasions, the use of spoken language is not conducive to communication and communication. For example, during machine translation, the input spoken text may not be accurately translated due to irregularities in the spoken language. [0004] But sometimes...

AI Technical Summary

Problems solved by technology

[0003] Because spoken language does not have the concise and standardized characteristics of written language, in some occasions, the use of spoken language is not conducive to communication and communication

For example, when performing machine translation, the input spoken text may not be translated accurately due to irregularities in the spoken language

Images