-
公开(公告)号：US06725197B1
公开(公告)日：2004-04-20
申请号：US09415832
申请日：1999-10-12
申请人： Friedhelm Wuppermann , Volker Stahl
发明人： Friedhelm Wuppermann , Volker Stahl
IPC分类号： G10L1504
CPC分类号： G10L15/22 , G10L2015/086
摘要： The invention relates to a method of automatic recognition of an at least partly spelled speech utterance, with a speech recognition unit (2) based on statistical models that include a linguistic speech model (6). To improve this method in such a way that in addition to a convenient entry, also a reduced speech recognition error rate is achieved, there is proposed that after the at least partly spelled speech utterance has been entered, the speech recognition unit (2) determines a first recognition result for the whole speech utterance; individually recognized letters are sent to the user for him to acknowledge or reject; after a letter has been acknowledged, the linguistic speech model (6) is adapted, which linguistic speech model, after its adaptation, determines the number of letters that can be allowed as followers of the acknowledged letters and assumes the correctness of letters already acknowledged; the speech recognition unit (2) determines a further recognition result for the speech utterance by means of the adapted linguistic speech model (6), from which result the next letter to be sent to the user is determined, so that he can acknowledge it.
摘要翻译： 本发明涉及一种基于包括语言语言模型（6）的统计模型的语音识别单元（2）自动识别至少部分拼写的语音发音的方法。为了改进这种方法，使得在 除了方便的输入外，还实现了降低的语音识别错误率，因此建议在输入了至少部分拼写的语音发音之后，语音识别单元（2）确定整个语音话语的第一识别结果;单独地 公认的信件被发送给用户以便他承认或拒绝;在一封信被确认之后，语言语言模型（6）被适应，哪个语言语言模型在其适应之后确定可以被允许的字母数 确认信的追随者并假定已经确认的字母的正确性;语音识别单元（2）确定语音的另外的识别结果 通过适应的语言语言模型（6）的话语，由此确定要发送给用户的下一个字母，以便他能够确认。
-
公开(公告)号：US08484582B2
公开(公告)日：2013-07-09
申请号：US12464756
申请日：2009-05-12
IPC分类号： G06F3/14
CPC分类号： G06F17/30716 , G06F17/30722
摘要： A method, device, and computer program product for locating a desired entry in a list containing multiple list entries for use with a limited text display is described. A list of entries is partitioned into a number of sub-parts such that the desired entry is contained in one of the list sub-parts. At least one of the sub-parts is characterized within a limited text display to prompt a user for feedback regarding the location of the desired entry in the sub-parts. The user feedback is received from a view input element. In response to the user feedback, the sub-part containing the desired entry is selected. The partitioning, characterizing, receiving, and selecting steps are re-performed one or more times until a final sub-part is generated that contains a limited number of entries including the desired entry.
摘要翻译： 描述了用于定位包含用于有限文本显示的多个列表条目的列表中的期望条目的方法，设备和计算机程序产品。 条目列表被划分成多个子部分，使得所需条目包含在列表子部分之一中。 至少一个子部件在有限的文本显示中被表征，以提示用户关于子部件中期望条目的位置的反馈。 从视图输入元件接收用户反馈。 响应于用户反馈，选择包含所需条目的子部分。 重新执行分区，表征，接收和选择步骤一次或多次，直到生成包含有限数量的条目（包括所需条目）的最终子部分。
-
公开(公告)号：US20100017393A1
公开(公告)日：2010-01-21
申请号：US12464756
申请日：2009-05-12
IPC分类号： G06F17/30
CPC分类号： G06F17/30716 , G06F17/30722
摘要： A method, device, and computer program product for locating a desired entry in a list containing multiple list entries for use with a limited text display is described. A list of entries is partitioned into a number of sub-parts such that the desired entry is contained in one of the list sub-parts. At least one of the sub-parts is characterized within a limited text display to prompt a user for feedback regarding the location of the desired entry in the sub-parts. The user feedback is received from a view input element. In response to the user feedback, the sub-part containing the desired entry is selected. The partitioning, characterizing, receiving, and selecting steps are re-performed one or more times until a final sub-part is generated that contains a limited number of entries including the desired entry.
摘要翻译： 描述了用于定位包含用于有限文本显示的多个列表条目的列表中的期望条目的方法，设备和计算机程序产品。 条目列表被划分成多个子部分，使得所需条目包含在列表子部分之一中。 至少一个子部件在有限的文本显示中被表征，以提示用户关于子部件中期望条目的位置的反馈。 从视图输入元件接收用户反馈。 响应于用户反馈，选择包含所需条目的子部分。 重新执行分区，表征，接收和选择步骤一次或多次，直到生成包含有限数量的条目（包括所需条目）的最终子部分。
-
公开(公告)号：US06272196B1
公开(公告)日：2001-08-07
申请号：US08798686
申请日：1997-02-12
IPC分类号： H04L2300
CPC分类号： G10L19/12 , G10L2019/0013
摘要： In a CELP coder, a comparison between a target signal and a plurality of synthetic signals is made. A synthetic signal is derived by filtering a plurality of excitation sequences by a synthesis filter having parameters derived from the target signal. The excitation signal, which results in a minimum error between the target signal and the synthetic signal, is selected. The search for the best excitation signal requires a substantial computational complexity. To reduce the complexity, a preselection of a small number of excitation sequences is made by selecting a small number of excitation sequences resembling the most backward filtered target signal. With this small number of excitation sequences a full complexity search is made. Due to the reduced number of excitation sequences involved in the final selection the required computational complexity is reduced.
摘要翻译： 在CELP编码器中，进行目标信号和多个合成信号之间的比较。 通过具有从目标信号导出的参数的合成滤波器对多个激励序列进行滤波而导出合成信号。 选择导致目标信号和合成信号之间的最小误差的激励信号。 搜索最佳激励信号需要大量的计算复杂度。 为了降低复杂度，通过选择少数类似于最后向滤波的目标信号的激励序列来进行少量激励序列的预选。 利用这种少量的激励序列，进行完全复杂的搜索。 由于在最终选择中涉及的激励序列数量减少，所需的计算复杂度降低。
-
公开(公告)号：US06600798B2
公开(公告)日：2003-07-29
申请号：US09761196
申请日：2001-01-16
IPC分类号： H04L2300
CPC分类号： G10L19/12 , G10L2019/0013
摘要： In a CELP coder, a comparison between a target signal and a plurality of synthetic signals is made. The synthetic signal is derived by filtering a plurality of excitation sequences by a synthesis filter having parameters derived from the target signal. The excitation signal which results in a minimum error between the target signal and synthetic signal is selected. The search for the best excitation signal requires a substantial computational complexity. To reduce the complexity, a preselection of a small number of excitation sequences is made using a reduced complexity synthesis filter. With this small number of excitation sequences, a full complexity search is made. Due to the reduced number of excitation sequences involved in the final selection, the required computational complexity is reduced.
摘要翻译： 在CELP编码器中，进行目标信号和多个合成信号之间的比较。 通过具有从目标信号导出的参数的合成滤波器对多个激励序列进行滤波而导出合成信号。 选择导致目标信号和合成信号之间的最小误差的激励信号。 搜索最佳激励信号需要大量的计算复杂度。 为了降低复杂度，使用降低复杂度的合成滤波器来进行少量激励序列的预选。 利用这种少量的激励序列，进行完全复杂的搜索。 由于最终选择中所涉及的激励序列数量减少，所需的计算复杂度降低。
-
公开(公告)号：US5982817A
公开(公告)日：1999-11-09
申请号：US962895
申请日：1997-11-03
申请人： Friedhelm Wuppermann
发明人： Friedhelm Wuppermann
IPC分类号： H04Q7/22 , G10L13/00 , G10L19/02 , G10L19/04 , G10L19/18 , H03M13/00 , H04B1/06 , H04B1/38 , H04B1/66 , H04B7/15 , H04B7/185 , H04B7/19 , H04B7/26 , H04B14/06 , H04Q7/24 , H04Q7/26 , H04Q7/30
CPC分类号： G10L19/04 , G10L19/0208 , G10L19/18 , H04B7/18578 , H04B7/19
摘要： In a transmission system for transmitting speech and music signals, an input signal is coded in a coder (11) by a time domain coder (4). The output signal of the time domain coder (4) is decoded by a time domain decoder (8) and the signal thus decoded is subtracted from the input signal by a subtracter circuit (10). To improve the coding quality, the difference signal is coded by a frequency domain coder (12) and the output signal of the time domain coder (4) and the frequency domain coder (12) are combined in a multiplexer (14) and transmitted to a receiver.
摘要翻译： 在用于发送语音和音乐信号的传输系统中，输入信号由编码器（11）由时域编码器（4）编码。 时域编码器（4）的输出信号由时域解码器（8）解码，并且由减法器电路（10）从输入信号中减去如此解码的信号。 为了提高编码质量，差分信号由频域编码器（12）编码，时域编码器（4）和频域编码器（12）的输出信号组合在多路复用器（14）中并发送到 接收器
-
公开(公告)号：US5808569A
公开(公告)日：1998-09-15
申请号：US320636
申请日：1994-10-11
CPC分类号： H04B1/667
摘要： A transmission system for encoded transmission of audio signals with frequency band compression. An input signal to be transmitted is divided into two spectral portions in a coder. One spectral portion is digitally encoded by a time-domain coder, and the other spectral portion is coded by a transform domain coder. The two digitally coded signals are then multiplexed for transmission as a composite digitally coded signal to a receiver. At the receiver the composite signal is demultiplexed and the two spectral portions thereof are respectively decoded by a time-domain decoder and a transformer-domain decoder. The decoded signals, covering the respective spectral portions, are then combined to reconstitute the originally transmitted audio signal.
摘要翻译： 用于频带压缩的音频信号的编码传输的传输系统。 要传输的输入信号在编码器中被分成两个频谱部分。 一个频谱部分由时域编码器进行数字编码，另一个频谱部分由变换域编码器编码。 然后将两个数字编码的信号复用以作为组合数字编码信号发送到接收机。 在接收机处，复合信号被解复用，并且其两个频谱部分分别由时域解码器和变压器域解码器解码。 然后将覆盖各个频谱部分的解码信号组合起来以重构原始传输的音频信号。
-
公开(公告)号：US06608877B1
公开(公告)日：2003-08-19
申请号：US09710820
申请日：2000-11-13
IPC分类号： H04L2300
CPC分类号： G10L19/12 , G10L2019/0013
摘要： In a CELP coder a comparison between a target signal and a plurality of synthetic signals is made. The synthetic signal is derived by filtering a plurality of excitation sequences by a synthesis filter having parameters derived from the target signal. The excitation signal which results in a minimum error between the target signal and the synthetic signal is selected. The search for the best excitation signal requires a substantial computational complexity. To reduce the complexity a preselection of a small number of excitation sequences is made by selecting a small number of excitation sequences resembling the most a backward filtered target signal. With this small number of excitation sequences a full complexity search is made. Due to the reduced number of excitation sequences involved in the final selection the required computational complexity is reduced.
摘要翻译： 在CELP编码器中，进行目标信号和多个合成信号之间的比较。 通过具有从目标信号导出的参数的合成滤波器对多个激励序列进行滤波而导出合成信号。 选择导致目标信号和合成信号之间的最小误差的激励信号。 搜索最佳激励信号需要大量的计算复杂度。 为了降低复杂度，通过选择少量的类似于反向滤波的目标信号的激励序列来进行少量激励序列的预选。 利用这种少量的激励序列，进行完全复杂的搜索。 由于在最终选择中涉及的激励序列数量减少，所需的计算复杂度降低。
-
9.发明授权CELP coding with two-stage search over displaced segments of a one-dimensional codebook 有权通过对一维码本的位移段进行两级搜索的CELP编码公开(公告)号：US06603832B2
公开(公告)日：2003-08-05
申请号：US09761197
申请日：2001-01-16
IPC分类号： H04L2300
CPC分类号： G10L19/12 , G10L2019/0013
摘要： In a CELP coder a comparison between a target signal and a plurality of synthetic signals is made. The synthetic signal is derived by filtering a plurality of excitation sequences from a one dimensional codebook by a synthesis filter having parameters derived from the target signal. The excitation signal that results in a minimum error between the target signal and the synthetic signal is selected. In order to reduce the complexity of the search for the best excitation signal, the selection is done in two stages. First a preselection of a small number of excitation sequences is made by selecting only every L.sup.th codebook entry for preselecting a plurality of excitation sequences. Thereafter, with this small number of excitation sequences, a fill complexity search is made in which all excitation sequences surrounding the preselected ones are involved in the selection.
摘要翻译： 在CELP编码器中，进行目标信号和多个合成信号之间的比较。 通过具有从目标信号导出的参数的合成滤波器对来自一维码本的多个激励序列进行滤波，得到合成信号。 选择导致目标信号与合成信号之间的最小误差的激励信号。 为了降低寻找最佳激励信号的复杂度，选择分两个阶段完成。 首先，通过仅选择每个L.sup.th码本条目来预选多个激励序列来进行少量激励序列的预选择。 此后，利用这种少量的激励序列，进行填充复杂度搜索，其中围绕预选的激励序列参与选择。
-
公开(公告)号：US6014619A
公开(公告)日：2000-01-11
申请号：US798889
申请日：1997-02-11
CPC分类号： G10L19/12
摘要： In a CELP coder a comparison between a target signal and a plurality of synthetic signals is made. The synthetic signal is derived by filtering a plurality of excitation sequences by a synthesis filter having parameters derived from the target signal. The excitation signal which results in a minimum error between the target signal and the synthetic signal is selected. In order to reduce the complexity of the search for the best excitation signal, thee selection is done in two stages. First a preselection of a small number of excitation sequences is made using a reduced complexity synthesis filter. Thereafter, with this small number of excitation sequences, a full complexity search is made. Due to the reduced number of excitation sequences involved in the final selection the required computational complexity is reduced.
摘要翻译： 在CELP编码器中，进行目标信号和多个合成信号之间的比较。 通过具有从目标信号导出的参数的合成滤波器对多个激励序列进行滤波而导出合成信号。 选择导致目标信号和合成信号之间的最小误差的激励信号。 为了降低寻找最佳激励信号的复杂度，选择分为两个阶段。 首先，使用降低复杂度的合成滤波器来进行少量激励序列的预选。 此后，利用这个少量的激励序列，进行完全复杂的搜索。 由于在最终选择中涉及的激励序列数量减少，所需的计算复杂度降低。
-
-
-
-
-
-
-
-
-
搜索结果
13 条记录 (耗时 0.017 秒)
