Browse Prior Art Database

An intellegent text to speech method based on character of roles

IP.com Disclosure Number: IPCOM000236145D
Publication Date: 2014-Apr-09
Document File: 5 page(s) / 80K

Publishing Venue

The IP.com Prior Art Database

Abstract

This idea provides a method to select appropriate voice for different roles which makes user be personally on the scene. It recognizes roles in novels and classifies them by characters. Each utterance of sentence they speak in novels will be matched to an appropriate voice. This method dubs different voices for roles in novels instead of using a changeless voice to make roles as vivid as life and improve user satisfaction.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 53% of the total text.

Page 01 of 5

An intellegent text to speech method based on character of roles

The text to speech(TTS) technology has been applied on many hand-on digital devices. Many users read text novels by TTS. In these devices/software there are some types of voice for speaker which permit user select one for text. But there are only several voices can be available to choose and the voice can't be changed during the speech.

http://blog.csdn.net/zhoubl668/article/details/7204322 http://www.cnblogs.com/eaglet/archive/2009/08/19/1549566.html http://wenku.baidu.com/view/91a3a00ef12d2af90242e65b.html

This idea provides a method to select appropriate voice for different roles which makes user be personally on the scene . It recognizes roles in novels and classifies them by characters. Each utterance of sentence they speak in novels will be matched to an appropriate voice. This method dubs different voices for roles in novels instead of using a changeless voice to make roles as vivid as life and improve user satisfaction.

The voices can be matched by following steps:


1. First, recognize the role names from text. The input text is received into the text-to-speech synthesizer system. The input text is subsequently analyzed to determine lexical and grammatical information. The lexical and grammatical analysis uses a variety of well known techniques. The role names can be recognized based on the frequency of occurrences in the text. And the user can correct them.


2. Recognize character of roles by semantic analyzer.(sex, age, character) and save into a library. Many well known semantic analysis techniques can be used to analyze the sentences and choose out the words which describe a specific subject, i.e., the role. For example, the semantic analyzer may use semantic analysis techniques such as symbolic machine learning, graph-based clustering and classification, statistics-based multivariate analyses, artificial neural network-based computing, or evolution-based programming. The characters are collected through the whole text.

For example, a role can be described as:

{XiaoXuan: Young,25 years old, Man,tall and slender}

...

...


3. Determine the speaker of the speech. In this step , the subjects are recognized by on well known semantic analyzer. The accuracy can be improved by following rules:

1



Page 02 of 5


1) Set a distance limit between speaker and speech.


2) The speaker can not be the names in the speech.

"Brother Lin, come on", the middle-aged man said.

speech speaker


3) If th...