Method and apparatus for modifying voice characteristics of synthesized speech
    2.
    发明公开
    Method and apparatus for modifying voice characteristics of synthesized speech 失效
    为合成产生的语音声音改变特性的方法和装置

    公开(公告)号:EP0762384A2

    公开(公告)日:1997-03-12

    申请号:EP96306091.8

    申请日:1996-08-21

    申请人: AT&T IPM Corp.

    IPC分类号: G10L5/04

    CPC分类号: G10L13/02 G10L13/04

    摘要: A speech synthesizer system and method is disclosed which facilitates experimentation with speech parameters that control acoustical characteristics of a base synthesized voice. The invention utilizes an interactive graphical user interface for manipulating and transmitting speech parameter values to a text to speech synthesizer. The text to speech synthesizer almost immediately converts test utterances, i.e., sample text to synthesize, to speech using the base synthesized voice altered according to the current speech parameter values received from the graphical user interface.

    摘要翻译: 一种语音合成系统和方法,是游离缺失光盘便于实验与语音参数做了控制的基座合成语音的声学特性。 本发明利用到交互式图形用户界面用于操纵和传输语音参数值,以文本到语音合成器。 所述文本到语音的合成器几乎立即使用基合成语音改变gemäß从图形用户界面接收的当前语音参数值测试话语,即,样本文本合成转换,到语音。

    System and method for voice controlled video screen display
    4.
    发明公开
    System and method for voice controlled video screen display 失效
    Verfahren und Einrichtung zur Sprachsteuerung von Bildschirmanzeigen

    公开(公告)号:EP0747881A2

    公开(公告)日:1996-12-11

    申请号:EP96303839.3

    申请日:1996-05-29

    申请人: AT&T IPM Corp.

    IPC分类号: G10L7/10 G10L7/08

    CPC分类号: H04M3/4938 G06F3/16 G10L15/26

    摘要: A system and method for a voice controlled video screen display system are provided. The voice controlled system is useful for providing "hands-free" navigation through various video screen displays such as the World Wide Web network and interactive television displays. During operation of the system, language models are derived from incoming data in applications such as the World Wide Web network.

    摘要翻译: 提供了一种用于语音控制的视频屏幕显示系统的系统和方法。 语音控制系统对于通过诸如万维网网络和交互式电视显示器的各种视频屏幕显示提供“免提”导航是有用的。 在系统运行期间,语言模型从诸如万维网网络的应用程序中的输入数据导出。