Text normalizing method and device

Abstract

The invention discloses a text normalizing method and device. The method comprises the steps of obtaining voice data and a recognition text corresponding to the voice data; obtaining a word unit in the recognition text and a word vector corresponding to the word unit; extracting regular features, specifically, extracting voice features from the voice data, and extracting text features from the recognition text; inputting the word vector and the feature vector corresponding to the structured feature into a pre-constructed text structured model, and obtaining a label corresponding to the word unit according to the output of the text structured model, the label at least comprising deletion; and normalizing the recognition text according to the label corresponding to the word unit to obtain anormalized speech recognition text. By utilizing the scheme of the invention, the spoken speech recognition text can be more understandable and can express normativity.

Description

technical field [0001] The invention relates to the field of text processing, in particular to a text regularization method and device. Background technique [0002] Simultaneous interpretation, referred to as "simultaneous interpretation", refers to the uninterrupted real-time translation of the speech content to the audience without interrupting the speaker. Its biggest feature is its high efficiency, and the audience can obtain information in a timely manner. It is widely used in important occasions such as international conferences and diplomatic negotiations. At present, with the development of artificial intelligence technology, machine simultaneous interpretation has appeared. The biggest advantage of machine simultaneous interpretation is that the translation rate will not decrease due to fatigue. However, because machine simultaneous interpretation is limited by speech recognition and machine translation technology, the overall translation quality of machine simult...

AI Technical Summary

Problems solved by technology

However, because machine simultaneous interpretation is limited by speech recognition and machine translation technology, the overall translation quality of machine simultaneous interpretation is far behind that of manual simultaneous interpretation. Especially in some speech occasions, speakers often improvise according to the theme and ideas. Speech, during the speech, there will be colloquial expressions such as incomplete semantics, repetition, and modal particles. This phenomenon will affect the quality of subsequent translations, and even lead to complete translation errors

In response to this situation, existing machine simultaneous interpretation products usually use a rule-based method to remove some colloquial expressions, but the effect is not good, which will greatly affect the effect of subsequent machine simultaneous interpretation

Images