公开(公告)号：US20140278417A1

公开(公告)日：2014-09-18

申请号：US13965661

申请日：2013-08-13

Applicant: Broadcom Corporation

Inventor： Juin-Hwey Chen ,  Robert W. Zopf ,  Bengt J. Borgstrom ,  Elias Nemer ,  Ashutosh Pandey ,  Jes Thyssen

IPC: G10L17/06 ,  G10L17/00

CPC classification number: G10L17/00 ,  G10L17/06 ,  G10L21/00

Abstract: Methods, systems, and apparatuses are described for performing speaker-identification-assisted speech processing. In accordance with certain embodiments, a communication device includes speaker identification (SID) logic that is configured to identify a user of the communication device and/or the identity of a far-end speaker participating in a voice call with a user of the communication device. Knowledge of the identity of the user and/or far-end speaker is then used to improve the performance of one or more speech processing algorithms implemented on the communication device.

Abstract translation: 描述了用于执行说话者识别辅助语音处理的方法，系统和装置。 根据某些实施例，通信设备包括说话人标识（SID）逻辑，其被配置为识别通信设备的用户和/或与通信设备的用户参与语音呼叫的远端扬声器的身份 。 然后使用对用户和/或远端扬声器的身份的知识来改善在通信设备上实现的一个或多个语音处理算法的性能。

公开(公告)号：US20130191120A1

公开(公告)日：2013-07-25

申请号：US13748949

申请日：2013-01-24

Applicant: Broadcom Corporation

Inventor： Robert W. Zopf ,  Juin-Hwey Chen

IPC: G10L19/005

CPC classification number: G10L19/005 ,  G10L19/12

Abstract: Methods, systems, and apparatuses for performing packet loss concealment are disclosed. In response to determining that an encoded frame representing a segment of a signal is bad, an encoded parameter within the encoded frame is decoded based on bit information (such as soft bit information) associated with the encoded parameter to obtain a decoded parameter. Whether the decoded parameter violates a parameter constraint is determined. If a parameter constraint violation is detected, an estimate of the decoded parameter is generated. Either the decoded parameter or estimate of the decoded parameter is passed to a decoder for use in decoding the encoded frame.

Abstract translation: 公开了用于执行分组丢失隐藏的方法，系统和装置。 响应于确定表示信号的段的编码帧是坏的，基于与编码参数相关联的比特信息（诸如软比特信息）来对编码帧内的编码参数进行解码以获得解码参数。 确定解码参数是否违反参数约束。 如果检测到参数约束违规，则生成解码参数的估计。 解码的参数或解码参数的估计被传递到解码器以用于解码编码帧。

公开(公告)号：US20160189706A1

公开(公告)日：2016-06-30

申请号：US14606588

申请日：2015-01-27

Applicant: Broadcom Corporation

Inventor： Robert W. Zopf ,  Bengt J. Borgstrom

IPC: G10L15/02 ,  G10L15/14 ,  G06F1/32 ,  G10L15/05

CPC classification number: G06F1/3231 ,  G10L15/063 ,  G10L15/144 ,  G10L2015/025 ,  G10L2015/0635 ,  G10L2015/088

Abstract: Methods, systems, and apparatuses are described for isolated word training and detection. Isolated word training devices and systems are provided in which a user may provide a wake-up phrase from 1 to 3 times to train the device or system. A concatenated phoneme model of the user-provided wake-up phrase may be generated based on the provided wake-up phrase and a pre-trained phoneme model database. A word model of the wake-up phrase may be subsequently generated from the concatenated phoneme model and the provided wake-up phrase. Once trained, the user-provided wake-up phrase may be used to unlock the device or system and/or to wake up the device or system from a standby mode of operation. The word model of the user-provided wake-up phrase may be further adapted based on additional provisioning of the wake-up phrase.

Abstract translation: 描述了用于孤立词训练和检测的方法，系统和装置。 提供了隔离词训练装置和系统，其中用户可以提供1到3次的唤醒短语来训练该装置或系统。 可以基于所提供的唤醒短语和预先训练的音素模型数据库来生成用户提供的唤醒短语的级联音素模型。 可以随后从连接的音素模型和所提供的唤醒短语中产生唤醒短语的单词模型。 经过训练后，用户提供的唤醒短语可用于解锁设备或系统和/或使设备或系统从待机操作模式唤醒。 用户提供的唤醒短语的单词模型可以基于唤醒短语的附加提供来进一步调整。

公开(公告)号：US09245531B2

公开(公告)日：2016-01-26

申请号：US14703512

申请日：2015-05-04

Applicant: Broadcom Corporation

Inventor： Robert W. Zopf

IPC: G10L19/008 ,  G10L19/002 ,  H04L1/00

CPC classification number: G10L19/008 ,  G10L19/00 ,  G10L19/002 ,  H03M13/2957 ,  H03M13/6306 ,  H03M13/6312 ,  H04J3/0632 ,  H04L1/005 ,  H04L1/0055 ,  H04L1/1812

Abstract: Methods, systems, and apparatuses are provided for performing joint source channel decoding in a manner that exploits parameter domain correlation. Redundancy in speech coding and packet field parameters is exploited to generate conditional probabilities that a decoder utilizes to perform joint source channel decoding. The conditional probabilities are based upon correlations of parameters of a current frame with parameters of the same or other frames or historical parameter data. Parameter domain correlation provides significant channel decoding improvement over prior bit domain solutions. Also provided are methods, systems, and apparatuses for utilizing received statistics of monitored data bits from which conditional probabilities are generated to perform channel decoding. The techniques described may be implemented at the decoder side and thus do not interfere with transmission standards.

Abstract translation: 提供了以利用参数域相关性的方式进行联合源信道解码的方法，系统和装置。 利用语音编码和分组字段参数的冗余来产生解码器利用来执行联合源信道解码的条件概率。 条件概率基于当前帧与相同或其他帧或历史参数数据的参数的参数的相关性。 参数域相关性提供了优于先前位域解决方案的重要信道解码改进。 还提供了用于利用从其产生条件概率来执行信道解码的受监视数据比特的接收统计的方法，系统和装置。 所描述的技术可以在解码器侧实现，因此不会干扰传输标准。

公开(公告)号：US20150235647A1

公开(公告)日：2015-08-20

申请号：US14703512

申请日：2015-05-04

Applicant: Broadcom Corporation

Inventor： Robert W. Zopf

IPC: G10L19/008 ,  G10L19/002

CPC classification number: G10L19/008 ,  G10L19/00 ,  G10L19/002 ,  H03M13/2957 ,  H03M13/6306 ,  H03M13/6312 ,  H04J3/0632 ,  H04L1/005 ,  H04L1/0055 ,  H04L1/1812

Abstract: Methods, systems, and apparatuses are provided for performing joint source channel decoding in a manner that exploits parameter domain correlation. Redundancy in speech coding and packet field parameters is exploited to generate conditional probabilities that a decoder utilizes to perform joint source channel decoding. The conditional probabilities are based upon correlations of parameters of a current frame with parameters of the same or other frames or historical parameter data. Parameter domain correlation provides significant channel decoding improvement over prior bit domain solutions. Also provided are methods, systems, and apparatuses for utilizing received statistics of monitored data bits from which conditional probabilities are generated to perform channel decoding. The techniques described may be implemented at the decoder side and thus do not interfere with transmission standards.

Abstract translation: 提供了以利用参数域相关性的方式进行联合源信道解码的方法，系统和装置。 利用语音编码和分组字段参数的冗余来产生解码器利用来执行联合源信道解码的条件概率。 条件概率基于当前帧与相同或其他帧或历史参数数据的参数的参数的相关性。 参数域相关性提供了优于先前位域解决方案的重要信道解码改进。 还提供了用于利用从其产生条件概率来执行信道解码的受监视数据比特的接收统计的方法，系统和装置。 所描述的技术可以在解码器侧实现，因此不会干扰传输标准。

公开(公告)号：US09037942B2

公开(公告)日：2015-05-19

申请号：US13749187

申请日：2013-01-24

Applicant: Broadcom Corporation

Inventor： Robert W. Zopf

IPC: H03M13/00 ,  H03M13/05 ,  H03M13/29

CPC classification number: H03M13/05 ,  H03M13/2957 ,  H03M13/6312 ,  H03M13/6525

Abstract: A turbo decoder is configured to perform iterative decoding of data bits of a data packet received via a source signal to converge on a “soft” decision representation for each data bit of the data packet. The turbo decoder includes both an interleaved decoder and a non-interleaved decoder that work collaboratively to refine and improve the “soft” decision of each of the originally-received data bits. The interleaved decoder and the non-interleaved decoder are injected with extrinsic information based on at least a-priori information of the source signal. The turbo decoder avoids positive feedback of the a-priori information regarding the source signal from one decoder to the other by subtracting out extrinsic information based on the a-priori information that is injected into a decoder from the “soft” decision(s) determined by the decoder.

Abstract translation: turbo解码器被配置为执行经由源信号接收的数据分组的数据比特的迭代解码，以收敛于数据分组的每个数据比特的“软”决策表示。 turbo解码器包括交织解码器和非交织解码器，其协同工作来改进和改进每个原始接收的数据比特的“软”决定。 基于源信号的至少先验信息，交织的解码器和非交错解码器被注入外部信息。 turbo解码器通过基于从所确定的“软”决定中注入到解码器中的先验信息减去外部信息，避免了关于源信号从一个解码器到另一个解码器的先验信息的正反馈 由解码器。

公开(公告)号：US08996962B2

公开(公告)日：2015-03-31

申请号：US13848624

申请日：2013-03-21

Applicant: Broadcom Corporation

Inventor： Robert W. Zopf ,  Prasanna Desai ,  Norbert Grunert

IPC: H03M13/00 ,  H03M13/05 ,  H03M13/45 ,  H04L9/06 ,  H03M13/09 ,  H03M13/11 ,  H03M13/15 ,  H03M13/23 ,  H03M13/29

CPC classification number: H03M13/05 ,  H03M13/09 ,  H03M13/1102 ,  H03M13/1515 ,  H03M13/152 ,  H03M13/23 ,  H03M13/2957 ,  H03M13/453 ,  H03M13/63 ,  H04L9/0656 ,  H04L2209/34

Abstract: A plurality of encrypted packets having common payload data are received, wherein each of the plurality of encrypted packets includes a corresponding parity check field, and wherein a corresponding parity check syndrome for each of the plurality of encrypted packets indicates at least one bit error. A payload portion of each of the plurality of encrypted packets is decrypted to generate a plurality of decrypted payload portions. At least one chase coding technique is used to generate a corrected decrypted payload, based on at least one candidate bit error position and further based on the corresponding parity check syndrome for at least one of the plurality of encrypted packets.

Abstract translation: 接收具有共同有效载荷数据的多个加密分组，其中多个加密分组中的每一个包括对应的奇偶校验字段，并且其中用于多个加密分组中的每一个的对应奇偶校验校验子码表示至少一个比特错误。 多个加密分组中的每一个的有效载荷部分被解密以产生多个解密的有效载荷部分。 至少一个追逐编码技术被用于基于至少一个候选比特错误位置生成校正的解密有效载荷，并且还基于用于多个加密分组中的至少一个的对应的奇偶校验校验子。

公开(公告)号：US20140032227A1

公开(公告)日：2014-01-30

申请号：US14042228

申请日：2013-09-30

Applicant: Broadcom Corporation

Inventor： Robert W. Zopf ,  Vivek Kumar ,  Michael S. Gustafson

IPC: G10L19/005

CPC classification number: G10L19/005 ,  H04L1/0045

Abstract: Systems and methods are described for managing bit errors present in an encoded bit stream representative of a portion of an audio signal, wherein the encoded bit stream is received via a channel in a wireless communications system. The channel may comprise, for example, a Synchronous Connection-Oriented (SCO) channel or an Extended SCO (eSCO) channel in a Bluetooth® wireless communications system.

Abstract translation: 描述了用于管理存在于表示音频信号的一部分的编码比特流中的比特错误的系统和方法，其中经由无线通信系统中的信道接收编码比特流。 该信道可以包括例如Bluetooth®无线通信系统中的同步面向连接（SCO）信道或扩展SCO（eSCO）信道。

公开(公告)号：US20130282366A1

公开(公告)日：2013-10-24

申请号：US13925616

申请日：2013-06-24

Applicant: BROADCOM CORPORATION

Inventor： Robert W. Zopf

IPC: G10L19/00

CPC classification number: G10L19/008 ,  G10L19/00 ,  G10L19/002 ,  H03M13/2957 ,  H03M13/6306 ,  H03M13/6312 ,  H04J3/0632 ,  H04L1/005 ,  H04L1/0055 ,  H04L1/1812

Abstract: Methods, systems, and apparatuses are provided for performing jitter buffer enhanced joint source channel decoding. Jitter buffer enhanced joint source channel decoding may be performed in a manner that exploits parameter domain correlation. A jitter buffer stores hard bits of properly channel decoded packets, and a secondary jitter buffer is implemented to store soft bits associated with packets that are improperly channel decoded. Joint source channel decoding may be delayed to perform channel decoding of a frame in the penultimate position of the jitter buffer. The soft bits stored in the secondary jitter buffer as well as hard bits stored in the jitter buffer, which may include future frames, are utilized to perform channel decoding. The delayed jitter buffer enhanced joint source channel decoding may also be extended to iteratively perform channel decoding for giving frames at each position in the jitter buffer as they traverse the jitter buffer.

Abstract translation: 提供了用于执行抖动缓冲增强联合源信道解码的方法，系统和装置。 可以以利用参数域相关性的方式执行抖动缓冲器增强的联合源信道解码。 抖动缓冲器存储适当的信道解码分组的硬比特，并且实现辅助抖动缓冲器以存储与不正确地信道解码的分组相关联的软比特。 可以延迟联合源信道解码以对抖动缓冲器的倒数第二个位置中的帧执行信道解码。 存储在辅助抖动缓冲器中的软比特位以及存储在抖动缓冲器中的可能包括未来帧的硬比特被用于执行信道解码。 延迟抖动缓冲器增强的联合源信道解码还可以被扩展以迭代地执行信道解码，以在抖动缓冲器中的每个位置处遍历抖动缓冲器时给出帧。

公开(公告)号：US20130191707A1

公开(公告)日：2013-07-25

申请号：US13749187

申请日：2013-01-24

Applicant: Broadcom Corporation

Inventor： Robert W. Zopf

IPC: H03M13/05

CPC classification number: H03M13/05 ,  H03M13/2957 ,  H03M13/6312 ,  H03M13/6525

Abstract: A turbo decoder is configured to perform iterative decoding of data bits of a data packet received via a source signal to converge on a “soft” decision representation for each data bit of the data packet. The turbo decoder includes both an interleaved decoder and a non-interleaved decoder that work collaboratively to refine and improve the “soft” decision of each of the originally-received data bits. The interleaved decoder and the non-interleaved decoder are injected with extrinsic information based on at least a-priori information of the source signal. The turbo decoder avoids positive feedback of the a-priori information regarding the source signal from one decoder to the other by subtracting out extrinsic information based on the a-priori information that is injected into a decoder from the “soft” decision(s) determined by the decoder.

Abstract translation: turbo解码器被配置为执行经由源信号接收的数据分组的数据比特的迭代解码，以收敛于数据分组的每个数据比特的“软”决策表示。 turbo解码器包括交织解码器和非交织解码器，其协同工作来改进和改进每个原始接收的数据比特的“软”决定。 基于源信号的至少先验信息，交织的解码器和非交错解码器被注入外部信息。 turbo解码器通过基于从所确定的“软”决定中注入到解码器中的先验信息减去外部信息，避免了关于源信号从一个解码器到另一个解码器的先验信息的正反馈 由解码器。