专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

US06963834B2 Method of speech recognition using empirically determined word candidates 有权
标题翻译：使用经验确定的词候选语音识别方法
公开(公告)号：US06963834B2
公开(公告)日：2005-11-08
申请号：US09867197
申请日：2001-05-29
申请人： Matthew W. Hartley , James R. Lewis , David E. Reich
发明人： Matthew W. Hartley , James R. Lewis , David E. Reich
IPC分类号： G10L15/08 , G10L15/22 , G10L15/26
CPC分类号： G10L15/08 , G10L2015/221
摘要： A method for performing speech recognition can include determining a recognition result for received user speech. The recognition result can include recognized text and a corresponding confidence score. The confidence score of the recognition result can correspond to a predetermined minimum threshold. If the confidence score does not exceed the predetermined minimum threshold, the user can be presented with at least one empirically determined alternate word candidate corresponding to the recognition result.
摘要翻译：用于执行语音识别的方法可以包括确定用于接收到的用户语音的识别结果。识别结果可以包括识别的文本和相应的置信度得分。识别结果的置信度分数可以对应于预定的最小阈值。如果置信度分数不超过预定的最小阈值，则可以向用户呈现与识别结果相对应的至少一个经验确定的替代词候选。

2. 发明授权

US06975986B2 Voice spelling in an audio-only interface 有权
标题翻译：仅限音频界面中的语音拼写
公开(公告)号：US06975986B2
公开(公告)日：2005-12-13
申请号：US09867832
申请日：2001-05-30
申请人： Matthew W. Hartley , James R. Lewis , David F. Ma
发明人： Matthew W. Hartley , James R. Lewis , David F. Ma
IPC分类号： G10L15/10 , G10L13/00 , G10L15/06 , G10L15/22 , G10L15/04
CPC分类号： G10L15/22 , G10L2015/225
摘要： A voice spelling method. A voice spelling method can include the steps of: in an audio-only interface, receiving a plurality of audio signals representative of spoken characters, the plurality of spoken characters specifying a string; and, through the audio-only interface, providing audible feedback in between each received spoken character. Additionally, the method can include the steps of: through the audio-only interface, audibly playing back each spoken character; accepting a voice selection of one of the played back characters, the selection denoting a disputed character; identifying a replacement character; and, replacing the disputed character with the identified replacement character in the specified string.
摘要翻译：语音拼写方式。语音拼写方法可以包括以下步骤：在仅音频接口中，接收表示口语字符的多个音频信号，所述多个口头字符指定字符串; 并且通过仅音频接口在每个接收到的口头角色之间提供可听见的反馈。此外，该方法可以包括以下步骤：通过仅音频接口，可听地回放每个口语角色; 接受一个播放的人物的声音选择，该选择表示有争议的角色; 识别替换字符; 并用指定字符串中标识的替换字符替换有争议的字符。

3. 发明授权

US06662157B1 Speech recognition system for database access through the use of data domain overloading of grammars 有权
标题翻译：语音识别系统通过使用数据库访问语法重写语法
公开(公告)号：US06662157B1
公开(公告)日：2003-12-09
申请号：US09596770
申请日：2000-06-19
申请人： Brenda Horowitz , James R. Lewis , David E. Reich
发明人： Brenda Horowitz , James R. Lewis , David E. Reich
IPC分类号： G10L1526
CPC分类号： G10L15/19
摘要： A method for voice data entry availability in a voice response system can include receiving speech input specifying data in an audio user interface to a data information system for processing data in a data store. The speech input can be received through an audio user interface to the data information system. Subsequently, speech-to-text conversion of the speech input can be performed using a speech recognition engine with reference to a corresponding speech grammar. In particular, the speech grammar can contain a data set of words relating to the data information system. Notably, the data store can contain a subset of the data set, the subset having words which can be processed by the data information system, the subset not having words which cannot be processed by the data information system. If the specified data is included in the speech grammar and if the specified data is in the data store, the speech data in the speech query can be processed. However, if the specified data is not in the data store, it can be reported that the specified data cannot be processed. Finally, if the specified data is not included in the speech grammar, an Out-Of-Grammar (OOG) condition can be reported. Additionally, the speech data in the speech query is not processed.
摘要翻译：用于话音响应系统中的话音数据输入可用性的方法可以包括在数据信息系统的音频用户界面中接收指定数据的语音输入，以处理数据存储中的数据。语音输入可以通过音频用户界面接收到数据信息系统。随后，可以使用语音识别引擎参考对应的语音语法来执行语音输入的语音到文本转换。特别地，语音语法可以包含与数据信息系统相关的单词的数据集。值得注意的是，数据存储可以包含数据集的子集，该子集具有可由数据信息系统处理的字，该子集不具有数据信息系统不能处理的字。如果指定的数据被包括在语音语法中，并且如果指定的数据在数据存储器中，则可以处理语音查询中的语音数据。但是，如果指定的数据不在数据存储中，则可以报告无法处理指定的数据。最后，如果指定的数据不包括在语音语法中，则可以报告语法外（OOG）。此外，语音查询中的语音数据未被处理。

4. 发明申请

US20080215316A1 METHOD AND SYSTEM FOR TRIMMING AUDIO FILES 失效
标题翻译：用于调整音频文件的方法和系统
公开(公告)号：US20080215316A1
公开(公告)日：2008-09-04
申请号：US12029343
申请日：2008-02-11
申请人： James R. Lewis , A. Longenberger Catherine , David E. Reich
发明人： James R. Lewis , A. Longenberger Catherine , David E. Reich
IPC分类号： G10L19/14
CPC分类号： G10L25/78 , G06F17/30746 , G06F17/30749 , G10L15/00
摘要： A system for automatically trimming an audio files based upon textual content associated with the audio file is provided. The source of the textual content may be an electronic document or written language text. The textual content may include predefined hints, a text mark, or end-of-phrase punctuation mark. The system generates a trimming instruction based upon textual content corresponding to the audio file, and the audio file is trimmed based upon the trimming instruction.
摘要翻译：提供了一种基于与音频文件相关联的文本内容来自动修整音频文件的系统。文本内容的来源可以是电子文档或书面语言文本。文本内容可以包括预定义的提示，文本标记或短语结尾的标点符号。该系统基于与音频文件对应的文本内容生成修剪指令，并且基于修剪指令修剪音频文件。

5. 发明授权

US07356469B2 Method and system for trimming audio files 失效
公开(公告)号：US07356469B2
公开(公告)日：2008-04-08
申请号：US10922760
申请日：2004-08-20
申请人： James R. Lewis , Catherine A. Longenberger , David E. Reich
发明人： James R. Lewis , Catherine A. Longenberger , David E. Reich
IPC分类号： G10L15/00
CPC分类号： G10L25/78 , G06F17/30746 , G06F17/30749 , G10L15/00
摘要： A system for automatically trimming an audio files based upon textual content associated with the audio file is provided. The source of the textual content may be an electronic document or written language text. The textual content may include predefined hints, a text mark, or end-of-phrase punctuation mark. The system generates a trimming instruction based upon textual content corresponding to the audio file, and the audio file is trimmed based upon the trimming instruction.

6. 发明申请

US20120209613A1 METHOD AND ARRANGEMENT FOR MANAGING GRAMMAR OPTIONS IN A GRAPHICAL CALLFLOW BUILDER 有权
标题翻译：用于管理图形呼叫建筑物中的灰度选项的方法和布置
公开(公告)号：US20120209613A1
公开(公告)日：2012-08-16
申请号：US13344193
申请日：2012-01-05
申请人： Ciprian Agapi , Felipe Gomez , James R. Lewis , Vanessa V. Michelini
发明人： Ciprian Agapi , Felipe Gomez , James R. Lewis , Vanessa V. Michelini
IPC分类号： G10L11/00
CPC分类号： G10L2015/228
摘要： A method (10) in a speech recognition application callflow can include the steps of assigning (11) an individual option and a pre-built grammar to a same prompt, treating (15) the individual option as a valid output of the pre-built grammar if the individual option is a potential valid match to a recognition phrase (12) or an annotation (13) in the pre-built grammar, and treating (14) the individual option as an independent grammar from the pre-built grammar if the individual option fails to be a potential valid match to the recognition phrase or the annotation in the pre-built grammar.
摘要翻译：语音识别应用程序调用流程中的方法（10）可以包括以下步骤：将单个选项和预先构建的语法分配给相同的提示，将（15）个别选项视为预先构建的有效输出如果个人选项是预先构建的语法中的识别短语（12）或注释（13）的潜在有效匹配，则将语法（14）作为独立语法从预先构建的语法处理（14），如果单个选项不能成为预先构建的语法中的识别短语或注释的潜在有效匹配。

7. 发明授权

US08102987B2 Adjusting music length to expected waiting time while caller is on hold 失效
标题翻译：调用者处于暂停状态时，将音乐长度调整为预期的等待时间
公开(公告)号：US08102987B2
公开(公告)日：2012-01-24
申请号：US12253164
申请日：2008-10-16
申请人： Ciprian Agapi , Thomas E. Creamer , James R. Lewis , Vanessa V. Michelini , Wallace J. Sadowski , Clifford J. Strohofer
发明人： Ciprian Agapi , Thomas E. Creamer , James R. Lewis , Vanessa V. Michelini , Wallace J. Sadowski , Clifford J. Strohofer
IPC分类号： H04M3/42
CPC分类号： H04M3/4285 , H04M3/5238
摘要： A method of adjusting music length to expected waiting time while a caller is on hold includes choosing one or more media selections based upon their play duration and matching the selection(s) to the expected waiting time.
摘要翻译：一种在呼叫者处于保持期间将音乐长度调整到预期等待时间的方法包括基于它们的播放持续时间选择一个或多个媒体选择，并将选择与期望的等待时间相匹配。

8. 发明授权

US07412378B2 Method and system of dynamically adjusting a speech output rate to match a speech input rate 有权
标题翻译：动态调整语音输出速率以匹配语音输入速率的方法和系统
公开(公告)号：US07412378B2
公开(公告)日：2008-08-12
申请号：US10815309
申请日：2004-04-01
申请人： James R. Lewis , Peeyush Jaiswal
发明人： James R. Lewis , Peeyush Jaiswal
IPC分类号： G10L21/00 , G10L13/00
CPC分类号： G10L21/04
摘要： A method (10) and system of adjusting a speech output rate to match a speech input rate can include the steps of receiving (12) speech input, computing (14) a speech input rate, and dynamically adjusting (18 or 26) a speech output rate to match the speech input rate. If the type of speech output is TTS, then a rate of TTS can be adjusted (18). If the type of speech output is recorded and alternate text is available, then steps (22 and 24) of counting alternate text available from a recorded output and determining an audio file length is used to compute a default output rate to adjust a recorded output rate. If the type is recorded and alternate text is unavailable, then steps (21 and 24) of obtaining an output word count from a transcription of a recorded speech output and determining an audio file length is used.
摘要翻译：一种方法（10）和调整语音输出速率以匹配语音输入速率的系统可以包括以下步骤：接收（12）语音输入，计算（14）语音输入速率，以及动态地调整（18或26）语音输出速率匹配语音输入速率。如果语音输出的类型是TTS，则可以调整TTS的速率（18）。如果语音输出的类型被记录并且替代文本可用，则使用计数从记录输出可用的替代文本并确定音频文件长度的步骤（22和24）来计算默认输出速率以调整记录的输出速率。如果记录类型并且替代文本不可用，则使用从记录的语音输出的转录获得输出字数并确定音频文件长度的步骤（21和24）。

9. 发明授权

US07072837B2 Method for processing initially recognized speech in a speech recognition session 有权
公开(公告)号：US07072837B2
公开(公告)日：2006-07-04
申请号：US09809904
申请日：2001-03-16
申请人： Kimberlee A. Kemble , James R. Lewis , Vanessa V. Michelini , Margarita Zabolotskaya
发明人： Kimberlee A. Kemble , James R. Lewis , Vanessa V. Michelini , Margarita Zabolotskaya
IPC分类号： G10L15/18
CPC分类号： G10L15/19 , G10L15/32
摘要： In a speech recognition session, a method of speech processing utilizing a plurality of grammars can include a series of steps. The method can include spawning a separate thread of execution corresponding to each one of a plurality grammars and allocating a post-processing recognition task to each separate thread of execution. Each post-processing recognition task can produce a possible recognition result using the corresponding grammar. A possible recognition result produced by one of the post-processing recognition tasks can be identified.

10. 发明授权

US06801897B2 Method of providing concise forms of natural commands 失效
标题翻译：提供简洁形式的自然命令的方法
公开(公告)号：US06801897B2
公开(公告)日：2004-10-05
申请号：US09819783
申请日：2001-03-28
申请人： Thomas A. Kist , James R. Lewis
发明人： Thomas A. Kist , James R. Lewis
IPC分类号： G10L1522
CPC分类号： G10L15/22 , G10L15/1822
摘要： In a speech recognition system, a method of providing a concise form of a natural command can include receiving a user specified command and identifying all functionally equivalent natural command forms of the user specified command from a natural command grammar. The length of each functionally equivalent natural command form can be determined. The method also can include presenting a selected one of the functionally equivalent natural command forms having the shortest length to the user.
摘要翻译：在语音识别系统中，提供自然命令的简洁形式的方法可以包括从自然命令语法接收用户指定的命令并且识别所述用户指定命令的所有功能等同的自然命令形式。可以确定每个功能等效的自然命令形式的长度。该方法还可以包括向用户呈现具有最短长度的功能等效的自然命令形式中的所选择的一个。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式