Method and system for enrolling addresses in a speech recognition
database
    1.
    发明授权
    Method and system for enrolling addresses in a speech recognition database 失效
    用于在语音识别数据库中注册地址的方法和系统

    公开(公告)号:US5832429A

    公开(公告)日:1998-11-03

    申请号:US712043

    申请日:1996-09-11

    IPC分类号: G10L15/06 G10L7/08

    CPC分类号: G10L15/063

    摘要: A method and system for enrolling speed dial names includes providing speaker dependent templates and associated telephone numbers and providing a penalized garbage model for unrecognized speech. When a request for a new template is received it is determined if the list of speed dial names is full (Step 201) and is not it is determined if that name is too similar (Step 205) to a name already on the speed dial list. If so, that name is rejected but if not it is determined if the speed dial name is too short (Step 302), and if not; too short or if the user wants to enter the short name the system asks the user to repeat the speed dial name and if a match it is entered. If not a match the system will swap the first and second utterance and compare to see if a match.

    摘要翻译: 用于登记快速拨号名称的方法和系统包括提供说话者依赖模板和相关联的电话号码,并为无法识别的语音提供惩罚垃圾模型。 当接收到对新模板的请求时,确定快速拨号名称的列表是否已满(步骤201),并且不确定该名称是否与已经在快速拨号列表上的名称太相似(步骤205) 。 如果是,该名称被拒绝,但如果不是,确定快速拨号名称是否太短(步骤302),如果不是, 如果用户想要输入短名称,系统会要求用户重复快速拨号名称,如果输入了匹配项。 如果不匹配,系统会交换第一个和第二个发音,并比较一下是否匹配。

    Speech recognition method and system with improved voice-activated
prompt interrupt capability
    3.
    发明授权
    Speech recognition method and system with improved voice-activated prompt interrupt capability 失效
    语音识别方法和系统具有改进的语音激活提示中断能力

    公开(公告)号:US5708704A

    公开(公告)日:1998-01-13

    申请号:US418327

    申请日:1995-04-07

    摘要: The present invention provides a system for a voice prompt talkover capability. The system and method include a speech recognizer (36), an echo canceler (24), a delay finder (27) and a utterance extraction mechanism (30). The delay finder (27) determines the propagation delay between outgoing signal and the incoming signal and stores the signal in a delay buffer (23) in accordance with the delay and the filter length of the echo canceler. The echo canceler (24), using the incoming signals, the delay buffer and the delay enhance the incoming signal by removing the echo. The utterance extractor (30) determines the start of an utterance based on energy in the vectors of the LPC buffer (28). Upon the detection of an utterance the utterance extractor (30) initiates the start of recognition and terminates operation of the echo canceler (24) and the prompt play out (14). A recognition result is returned from the speech recognition processor (36) to the application (12).

    摘要翻译: 本发明提供了一种用于语音提示聊天能力的系统。 该系统和方法包括语音识别器(36),回波消除器(24),延迟取景器(27)和话音提取机构(30)。 延迟取景器(27)确定输出信号和输入信号之间的传播延迟,并根据回波消除器的延迟和滤波器长度将信号存储在延迟缓冲器(23)中。 使用输入信号的回波消除器(24)延迟缓冲器和延迟通过去除回波来增强输入信号。 发声提取器(30)基于LPC缓冲器(28)的矢量中的能量来确定发音的开始。 当检测到话语时,话音提取器(30)启动识别开始并终止回波消除器(24)的操作和迅速播出(14)。 识别结果从语音识别处理器(36)返回到应用程序(12)。