Speech synthesis employing concatenated prosodic and acoustic templates for phrases of multiple words
    21.
    发明授权
    Speech synthesis employing concatenated prosodic and acoustic templates for phrases of multiple words 有权
    语音合成采用多个词短语的连接韵律和声学模板

    公开(公告)号:US06496801B1

    公开(公告)日:2002-12-17

    申请号:US09432876

    申请日:1999-11-02

    IPC分类号: G10L1308

    CPC分类号: G10L13/08

    摘要: A speech synthesis system for generating voice dialog for a message frame having a fixed and a variable portion. A prosody module selects a prosodic template for each of the fixed and variable portions wherein at least one portion comprises a phrase of multiple words. An acoustic module selects an acoustic template for each of the fixed and variable portions wherein at least one portion comprises a phrase of multiple words. A frame generator concatenates the respective prosodic templates and acoustic templates. A sound module generates the voice dialog in accordance with the concatenated prosodic and acoustic templates.

    摘要翻译: 一种用于产生具有固定和可变部分的消息帧的语音对话的语音合成系统。 韵律模块为每个固定和可变部分选择韵律模板,其中至少一个部分包括多个词的短语。 声学模块为每个固定和可变部分选择声学模板,其中至少一个部分包括多个单词的短语。 帧发生器连接相应的韵律模板和声学模板。 声音模块根据连接的韵律和声学模板产生语音对话。

    Identification of unit overlap regions for concatenative speech synthesis system
    22.
    发明授权
    Identification of unit overlap regions for concatenative speech synthesis system 失效
    串联语音合成系统单元重叠区域的识别

    公开(公告)号:US06202049B1

    公开(公告)日:2001-03-13

    申请号:US09264981

    申请日:1999-03-09

    IPC分类号: G10L1306

    CPC分类号: G10L13/07

    摘要: Speech signal parameters are extracted from time-series data corresponding to different sound units containing the same vowel. The extracted parameters are used to train a statistical model, such as a Hidden Markov-based Model, that has a data structure for separately modeling the nuclear trajectory region of the vowel and its surrounding transition elements. The model is trained as through embedded re-estimation to automatically determine optimally aligned models that identify the nuclear trajectory region. The boundaries of the nuclear trajectory region serve to delimit the overlap region for subsequent sound unit concatenation.

    摘要翻译: 从与包含相同元音的不同声音单元对应的时间序列数据中提取语音信号参数。 提取的参数用于训练诸如基于隐马尔可夫模型的统计模型,其具有用于分别建模元音的核轨迹区域及其周围过渡元素的数据结构。 该模型通过嵌入式重估进行训练,以自动确定识别核轨迹区域的最佳对齐模型。 核轨迹区域的边界用于限定后续声音单元连接的重叠区域。

    User interface controller for text-to-speech synthesizer
    23.
    发明授权
    User interface controller for text-to-speech synthesizer 失效
    用于文本到语音合成器的用户界面控制器

    公开(公告)号:US5850629A

    公开(公告)日:1998-12-15

    申请号:US709582

    申请日:1996-09-09

    CPC分类号: G10L13/08 G06F3/16 G10L13/04

    摘要: An application-independent, text-to-speech control system which employs an easy-to-use transport from which a user can control most text-to-speech conversion functions without prior training. It provides a control means by which user designated areas may be sequentially selected, stored and played. Furthermore, moving back and forward through user-selected text may be controlled by the user, with the synthesized speech output maintaining the proper pauses and inflections.

    摘要翻译: 一种独立于应用的文本到语音控制系统,其采用易于使用的传输,用户可以在此之前控制大多数文本转语音转换功能,而无需事先进行培训。 它提供了一个控制装置,通过该控制装置可以依次选择,存储和播放用户指定的区域。 此外,通过用户选择的文本返回和前进可以由用户控制,合成语音输出保持适当的暂停和变形。

    Recovering multi-volume data sets during volume recovery
    24.
    发明授权
    Recovering multi-volume data sets during volume recovery 失效
    在卷恢复期间恢复多卷数据集

    公开(公告)号:US5644696A

    公开(公告)日:1997-07-01

    申请号:US470732

    申请日:1995-06-06

    CPC分类号: G06F11/1469

    摘要: A system for recovering multi-volume data sets that are stored in one or more volumes that have failed with a substantial level of automation and flexibility. Data can be recovered from backup storage when multiple volumes fail, as in a "catastrophic" data loss. For each volume, all data sets partially or completely stored on the selected volume prior to a failure of that volume are identified. Each single-volume data set stored on the selected volume prior to the failure is restored from backup storage. Upon identifying a multi-volume data set whose beginning segment was stored on the selected volume, the entire data set is restored. Data can also be recovered from backup storage when a particular volume fails, such as after a "head crash." All data sets partially or completely stored on the selected volume prior to a failure of that volume are identified. Each single volume data set stored on the selected volume prior to the failure is recovered from backup storage. For each multi-volume data set having a segment stored on the selected volume prior to the failure, a predetermined message is issued to a human operator. Alternatively, multi-volume data sets are recovered if consistent. Otherwise, if valid data has been lost, a warning message is issued.

    摘要翻译: 一种用于恢复多卷数据集的系统,该系统存储在一个或多个卷中,具有大量自动化和灵活性的失败。 当多个卷失败时,数据可以从备份存储中恢复,如“灾难性”的数据丢失。 对于每个卷,识别在该卷故障之前部分地或完全地存储在所选择的卷上的所有数据集。 存储在故障之前的所选卷上的每个单卷数据集从备份存储中恢复。 在识别出开始段存储在所选择的卷上的多卷数据集时,整个数据集被恢复。 数据也可以在特定卷失败时从备份存储中恢复,例如在“头撞”之后。 识别在该卷故障之前部分地或完全地存储在所选择的卷上的所有数据集。 存储在故障之前所选卷上的每个单个卷数据集从备份存储器中恢复。 对于在故障之前具有存储在所选择的卷上的段的每个多卷数据集,向人类操作者发出预定消息。 或者,如果一致,则恢复多卷数据集。 否则,如果有效数据丢失,则会发出警告消息。