A method and system for document speech processing based on intelligent indexing

Abstract

The invention provides a file phoneticization processing method based on intelligent indexing, which comprises the following steps of: indexing original text blocks and picture blocks which are extracted from a digital file so as to combine the text blocks, calibrate the content property of the text blocks, associate the picture blocks with illustrated text blocks and construct an article; establishing a text information list which describes a data relation among different articles and / or text contents in the same article, an associating relation between the picture blocks and the illustratedtext blocks and / or picture information and a reading sequence; and transmitting information of the text information list into a voice library to generate a voice record file or product or read with voices. Correspondingly, the invention also provides a file phoneticization processing system. By the method and the system, different phoneticization modes of different text blocks and paragraph switching during phoneticization are realized, the flexibility and the variety of the phoneticization are improved, and the reading interest is enhanced; furthermore, a user can preset the reading sequenceand voice library parameters, so the use flexibility is high.

Description

technical field [0001] The invention belongs to the technical field of information processing, and in particular relates to a method and system for voice processing of documents based on intelligent indexing. Background technique [0002] With the development of voice technology, voice reading has become an important function on various terminal devices, providing terminal users with auditory enjoyment and bringing a new reading experience. However, currently available voice reading software and published document voice reading methods basically read page by page without identifying and distinguishing content, and the method is single. Even some voice software that supports drag-and-drop reading can only be realized under manual intervention. For example, Free Read software requires the user to manually select some text to achieve drag-and-drop reading. For the user, the flexibility is poor and the method is single. [0003] In the Chinese patent application "A System and...

AI Technical Summary

Problems solved by technology

This method has the following problems: 1. Only distinguishing the attributes of different text contents according to the semantics cannot effectively and correctly identify the attributes of the text contents in digital files with complex layouts; Recognition, therefore, cannot be applied to voice reading of digital files containing pictures, especially for digital files with rich information expressed in pictures, the lack of pictures will inevitably lead to the loss of a large amount of information, resulting in incoherent or wrong reading content; 3. The The method allows users to subscribe to different text content in text files, but it cannot realize paragraph jumps during the reading process, and users cannot set different reading methods and reading orders for different content according to their own needs, so the flexibility of use is poor

Images