Training A Transcription System
    1.
    发明申请
    Training A Transcription System 有权
    培训转录系统

    公开(公告)号:US20110276325A1

    公开(公告)日:2011-11-10

    申请号:US12774054

    申请日:2010-05-05

    IPC分类号: G10L15/26

    CPC分类号: G10L15/07 G10L2015/0638

    摘要: According to certain embodiments, training a transcription system includes accessing recorded voice data of a user from one or more sources. The recorded voice data comprises voice samples. A transcript of the recorded voice data is accessed. The transcript comprises text representing one or more words of each voice sample. The transcript and the recorded voice data are provided to a transcription system to generate a voice profile for the user. The voice profile comprises information used to convert a voice sample to corresponding text.

    摘要翻译: 根据某些实施例,训练转录系统包括从一个或多个来源访问用户的记录语音数据。 记录的语音数据包括语音样本。 录制的语音数据的记录被访问。 成绩单包括表示每个语音样本的一个或多个单词的文本。 将转录本和记录的语音数据提供给转录系统以生成用户的语音简档。 语音简档包括用于将语音样本转换为相应文本的信息。

    Training a transcription system
    2.
    发明授权
    Training a transcription system 有权
    训练转录系统

    公开(公告)号:US09009040B2

    公开(公告)日:2015-04-14

    申请号:US12774054

    申请日:2010-05-05

    IPC分类号: G10L15/00 G10L15/07 G10L15/06

    CPC分类号: G10L15/07 G10L2015/0638

    摘要: According to certain embodiments, training a transcription system includes accessing recorded voice data of a user from one or more sources. The recorded voice data comprises voice samples. A transcript of the recorded voice data is accessed. The transcript comprises text representing one or more words of each voice sample. The transcript and the recorded voice data are provided to a transcription system to generate a voice profile for the user. The voice profile comprises information used to convert a voice sample to corresponding text.

    摘要翻译: 根据某些实施例,训练转录系统包括从一个或多个来源访问用户的记录语音数据。 记录的语音数据包括语音样本。 录制的语音数据的记录被访问。 成绩单包括表示每个语音样本的一个或多个单词的文本。 将转录本和记录的语音数据提供给转录系统以生成用户的语音简档。 语音简档包括用于将语音样本转换为相应文本的信息。

    Speaker-cluster dependent speaker recognition (speaker-type automated speech recognition)
    3.
    发明授权
    Speaker-cluster dependent speaker recognition (speaker-type automated speech recognition) 有权
    扬声器群集相关扬声器识别(扬声器型自动语音识别)

    公开(公告)号:US08600750B2

    公开(公告)日:2013-12-03

    申请号:US12795959

    申请日:2010-06-08

    IPC分类号: G10L15/06

    CPC分类号: G10L15/075 G10L2015/0631

    摘要: In an example embodiment, there is disclosed herein an automatic speech recognition (ASR) system that employs speaker clustering (or speaker type) for transcribing audio. A large corpus of audio with corresponding transcripts is analyzed to determine a plurality of speaker types (e.g., dialects). The ASR system is trained for each speaker type. Upon encountering a new user, the ASR system attempts to map the user to a speaker type. After the new user is mapped to a speaker type, the ASR employs the speaker type for transcribing audio from the new user.

    摘要翻译: 在示例实施例中,这里公开了一种使用用于转录音频的扬声器聚类(或扬声器类型)的自动语音识别(ASR)系统。 分析具有对应的抄本的大的音频语料库以确定多个扬声器类型(例如,方言)。 针对每个扬声器类型训练ASR系统。 在遇到新用户时,ASR系统尝试将用户映射到扬声器类型。 在新用户映射到扬声器类型之后,ASR使用扬声器类型从新用户转录音频。

    SPEAKER-CLUSTER DEPENDENT SPEAKER RECOGNITION (SPEAKER-TYPE AUTOMATED SPEECH RECOGNITION)
    4.
    发明申请
    SPEAKER-CLUSTER DEPENDENT SPEAKER RECOGNITION (SPEAKER-TYPE AUTOMATED SPEECH RECOGNITION) 有权
    扬声器相关扬声器识别(扬声器型自动语音识别)

    公开(公告)号:US20110301949A1

    公开(公告)日:2011-12-08

    申请号:US12795959

    申请日:2010-06-08

    IPC分类号: G10L15/00

    CPC分类号: G10L15/075 G10L2015/0631

    摘要: In an example embodiment, there is disclosed herein an automatic speech recognition (ASR) system that employs speaker clustering (or speaker type) for transcribing audio. A large corpus of audio with corresponding transcripts is analyzed to determine a plurality of speaker types (e.g., dialects). The ASR system is trained for each speaker type. Upon encountering a new user, the ASR system attempts to map the user to a speaker type. After the new user is mapped to a speaker type, the ASR employs the speaker type for transcribing audio from the new user.

    摘要翻译: 在示例实施例中,这里公开了一种使用用于转录音频的扬声器聚类(或扬声器类型)的自动语音识别(ASR)系统。 分析具有对应的抄本的大的音频语料库以确定多个扬声器类型(例如,方言)。 针对每个扬声器类型训练ASR系统。 在遇到新用户时,ASR系统尝试将用户映射到扬声器类型。 在新用户映射到扬声器类型之后,ASR使用扬声器类型从新用户转录音频。

    Reconferencing capability for networked phones
    6.
    发明授权
    Reconferencing capability for networked phones 有权
    网络电话的会议功能

    公开(公告)号:US09094520B2

    公开(公告)日:2015-07-28

    申请号:US11620943

    申请日:2007-01-08

    IPC分类号: H04M3/42 H04M3/56

    CPC分类号: H04M3/56 H04M2203/2088

    摘要: In one embodiment, a method comprises storing information associated with a conference call that has previously been terminated. The method also comprises obtaining a request to re-establish the terminated conference call, and re-establishing the conference call in response to the request to re-establish the conference call. Re-establishing the conference call comprises using the stored information to set a first set of states associated with the re-established conference call.

    摘要翻译: 在一个实施例中,一种方法包括存储与先前已被终止的电话会议相关联的信息。 该方法还包括获得重新建立终止的电话会议的请求,以及响应于重新建立会议呼叫的请求重新建立会议呼叫。 重新建立电话会议包括使用所存储的信息来设置与重新建立的电话会议相关联的第一组状态。

    Methods for ectopy rejection for atrial fibrillation detection based on ventricular cycle lengths
    7.
    发明授权
    Methods for ectopy rejection for atrial fibrillation detection based on ventricular cycle lengths 有权
    基于心室周期长度的心房颤动检测的异位排除方法

    公开(公告)号:US08977350B2

    公开(公告)日:2015-03-10

    申请号:US13050091

    申请日:2011-03-17

    摘要: A medical device performs a method to classify a cardiac rhythm. Differences between cycle lengths in a first heart chamber are determined during an established time interval. Evidence of ectopy associated with irregular coupling intervals is detected from the signal during the established time interval. A rhythm classification output corresponding to a second heart chamber at the expiration of the established time interval is provided in response to the consecutive cycle length differences and the evidence of ectopy associated with irregular coupling intervals.

    摘要翻译: 医疗器械执行心律的分类方法。 在建立的时间间隔期间确定第一心室中的循环长度之间的差异。 在建立的时间间隔期间从信号中检测到与不规则耦合间隔相关的异常的证据。 响应于连续循环长度差异和与不规则耦合间隔相关联的异常的证据,提供对应于在建立的时间间隔期满时的第二心室的节律分类输出。

    Geographical location information/signal quality-context based recording and playback of multimedia data from a conference session
    8.
    发明授权
    Geographical location information/signal quality-context based recording and playback of multimedia data from a conference session 有权
    地理位置信息/信号质量上下文记录和回放来自会议会话的多媒体数据

    公开(公告)号:US08938216B2

    公开(公告)日:2015-01-20

    申请号:US12953607

    申请日:2010-11-24

    IPC分类号: H04M1/725 H04N9/82

    摘要: Playback of multimedia data associated with a recorded conference session is simplified, particularly for a person who participated in the conference session from a mobile device. At a recording server that is in communication with a plurality of endpoints in a conference session, multimedia received from the plurality of endpoints participating in the conference session is recorded. The recording server obtains geographical location information for at least one mobile endpoint participating in the conference session as the at least one mobile endpoint moves. The recording server temporally correlates the geographical location information for the at least one endpoint with recorded multimedia data for the conference session. The recording server stores data tags for the geographical location information for the at least one mobile endpoint in association with the recorded multimedia data for subsequent use in indexing to map data to retrieve a portion of the recorded multimedia data for playback.

    摘要翻译: 与记录的会议会话相关联的多媒体数据的回放被简化,特别是对于从移动设备参与会议会话的人来说。 在与会议会话中的多个端点通信的记录服务器中,记录从参与会议会话的多个端点接收到的多媒体。 当至少一个移动终端移动时,记录服务器获得参与会议会话的至少一个移动终端的地理位置信息。 记录服务器将至少一个端点的地理位置信息与会议会话的记录的多媒体数据暂时相关。 记录服务器存储与记录的多媒体数据相关联的用于至少一个移动端点的地理位置信息的数据标签,用于随后用于索引以映射数据以检索记录的多媒体数据的一部分用于回放。

    Method and system for recording automatic call distributor calls
    9.
    发明授权
    Method and system for recording automatic call distributor calls 有权
    记录自动呼叫分配器呼叫的方法和系统

    公开(公告)号:US08488775B2

    公开(公告)日:2013-07-16

    申请号:US12962153

    申请日:2010-12-07

    IPC分类号: H04M3/00

    摘要: A method for recording automatic call distributor calls includes receiving from a caller a call for distribution to one of a plurality of agents and distributing the call to a first agent of the plurality of agents for handling. The method includes creating a call signature associated with the call for recording verification and recording communications of the call to a recording file. The method also includes embedding the call signature in the recording file.

    摘要翻译: 用于记录自动呼叫分配器呼叫的方法包括从呼叫者接收分配给多个代理之一的呼叫并将呼叫分发给多个代理的第一代理进行处理。 该方法包括创建与用于记录验证的呼叫相关联的呼叫签名并将呼叫的通信记录到记录文件。 该方法还包括将呼叫签名嵌入到记录文件中。

    Quantifying autonomic tone with thoracic impedance
    10.
    发明授权
    Quantifying autonomic tone with thoracic impedance 有权
    量化自主神经与胸阻抗

    公开(公告)号:US08435186B2

    公开(公告)日:2013-05-07

    申请号:US12696753

    申请日:2010-01-29

    IPC分类号: A61B5/00

    摘要: The disclosure describes techniques for quantifying the autonomic nervous system (ANS) health of a patient with thoracic impedance measurements. Thoracic impedance may be measured utilizing cardiac electrodes and an implantable medical device housing or other electrodes located on or within the patient. Since greater variability in thoracic impedance may indicate better health of the ANS, monitoring impedance changes in a patient may be used to quantify autonomic tone of the ANS, and ultimately, overall patient health. In some examples, thoracic impedance may be measured in response to a change in patient posture for acute monitoring or at predetermined times over several days, weeks, or months for more chronic monitoring of the patient. An implantable medical device may independently analyze the impedance measurements and transmit an alert to the patient or clinician when impedance changes indicate a change in patient health.

    摘要翻译: 本公开描述了用于量化具有胸部阻抗测量的患者的自主神经系统(ANS)健康的技术。 可以使用心电极和位于患者内部或内部的可植入医疗装置外壳或其它电极来测量胸腔阻抗。 由于胸部阻抗变化较大可能表明ANS的健康状况良好,所以监测患者的阻抗变化可用于量化ANS的自主音调,并最终对整体患者的健康状况进行量化。 在一些示例中,可以响应于用于急性监测的患者姿势的变化或在几天,几周或几个月的预定时间来测量胸部阻抗以进行更长期的患者监测。 当阻抗变化表示患者健康状况发生变化时,植入式医疗设备可以独立地分析阻抗测量结果并向患者或临床医生发送警报。