首页> 外国专利> Manner and the device which train the time long forecasting model, time long forecasting manner and the device and speech synthesis manner and the device

Manner and the device which train the time long forecasting model, time long forecasting manner and the device and speech synthesis manner and the device

机译:训练时间长预测模型的方式和装置,时间长预测方式和装置以及语音合成方式和装置

摘要

The present invention provides a method and apparatus for training a duration prediction model, method and apparatus for duration prediction and method and apparatus for speech synthesis. Said method for training a duration prediction model and comprising: generating an initial duration prediction model with a plurality of attributes related to duration prediction and at least part of possible attribute combinations of said plurality of attributes and in which each of said plurality of attributes and said attribute combinations is included as an item; calculating importance of each said item in said duration prediction model; deleting the item having the lowest importance calculated; re-generating a duration prediction model with the remaining items; determining whether said re-generated duration prediction model is an optimal model; and repeating said step of calculating importance and the following steps and if said duration prediction model is determined as not optimal model.
机译:本发明提供了用于训练持续时间预测模型的方法和设备,用于持续时间预测的方法和设备以及用于语音合成的方法和设备。所述用于训练持续时间预测模型的方法,并且包括:生成初始持续时间预测模型,该初始持续时间预测模型具有与持续时间预测有关的多个属性以及所述多个属性的可能属性组合的至少一部分,并且其中所述多个属性中的每一个和所述属性组合作为项目包括在内;计算所述持续时间预测模型中每个所述项目的重要性;删除具有最低重要性的项目;用剩余的项目重新生成持续时间预测模型;确定所述重新生成的持续时间预测模型是否是最佳模型;如果确定所述持续时间预测模型不是最佳模型,则重复所述计算重要性的步骤和随后的步骤。

著录项

  • 公开/公告号JP4602307B2

    专利类型

  • 公开/公告日2010-12-22

    原文格式PDF

  • 申请/专利权人 株式会社東芝;

    申请/专利号JP20060286910

  • 发明设计人 ジー・ハオ;イー・リフ;

    申请日2006-10-20

  • 分类号G10L13/08;

  • 国家 JP

  • 入库时间 2022-08-21 18:17:18

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号