公开(公告)号：WO0031720A2

公开(公告)日：2000-06-02

申请号：PCT/SE9902073

申请日：1999-11-12

Applicant: ERICSSON TELEFON AB L M

Inventor： SVEDBERG JONAS ,  EKUDDEN ERIK ,  UVLIDEN ANDERS ,  JOHANSSON INGEMAR

IPC: G10L11/02 ,  G10L15/00 ,  G10L

CPC classification number: G10L25/78 ,  G10L2025/783

Abstract: Perceptually relevant non-speech information can be preserved during encoding of an audio signal by determining whether the audio signal includes such information (122, 124, 125). If so, a speech/noise classification of the audio signal is overriden (43) to prevent misclassification of the audio signal as noise.

Abstract translation: 通过确定音频信号是否包括这样的信息（122,124,125）可以在编码音频信号期间保留感知相关的非语音信息。 如果是这样，音频信号的语音/噪声分类被覆盖（43），以防止音频信号被错误地分类为噪声。

公开(公告)号：WO0031719A3

公开(公告)日：2003-03-20

申请号：PCT/SE9902023

申请日：1999-11-08

Applicant: ERICSSON TELEFON AB L M

Inventor： EKUDDEN ERIK ,  HAGEN ROAR ,  JOHANSSON INGEMAR

IPC: G10L19/00 ,  H04B14/04 ,  H04M1/00 ,  H04M1/725

CPC classification number: G10L19/012

Abstract: The quality of comfort noise generated by a speech decoder (93) during non-speech periods is improved by modifying (30, 75) comfort noise parameter values (33) normally used to generate the comfort noise. The comfort noise parameter values are modified in response to variability information (43) associated with a background noise parameter. The modified comfort noise parameter values (35) are then used to generate the comfort noise.

Abstract translation: 通过修改（30,75）通常用于产生舒适噪声的舒适噪声参数值（33）来提高在非语音周期期间由语音解码器（93）产生的舒适噪声的质量。 响应于与背景噪声参数相关联的可变性信息（43）来修改舒适噪声参数值。 改进的舒适噪声参数值（35）然后用于产生舒适噪声。

公开(公告)号：WO0228075A3

公开(公告)日：2002-09-26

申请号：PCT/SE0102099

申请日：2001-09-27

Applicant: ERICSSON TELEFON AB L M

Inventor： EKUDDEN ERIK ,  ERIKSSON ANDERS

IPC: H04M3/22 ,  H04M3/56 ,  H04R3/00

CPC classification number: H04M3/56 ,  H04M3/2227 ,  H04M3/2236 ,  H04M3/568 ,  H04R27/00

Abstract: Cross talk terms may be removed from speech signals in a multi-party conference system having a plurality of users. In a preferred embodiment of the present invention, an estimator (210, 310) is used to first estimate the respective cross talk terms for each of the said plurality of users. These terms are then multiplied with a filter (230, 330) based on the estimated cross talk terms. Source speech signals are thereby recovered from the plurality of multi-party conference users without the cross talk.

Abstract translation: 可以在具有多个用户的多方会议系统中从语音信号中去除串扰词。 在本发明的优选实施例中，使用估计器（210,310）来首先估计所述多个用户中的每一个的相应串扰项。 然后，基于所估计的串扰项，将这些项与滤波器（230,330）相乘。 源语音信号从而从多个多方会议用户中恢复，而没有串扰。

公开(公告)号：WO0163774A8

公开(公告)日：2002-03-28

申请号：PCT/SE0100394

申请日：2001-02-22

Applicant: ERICSSON TELEFON AB L M

Inventor： EKUDDEN ERIK ,  SJOEBERG JOHAN

IPC: H03M13/35 ,  G10L21/00

CPC classification number: H03M13/35

Abstract: A method and apparatus for partial redundancy encoding of a speech data packet is disclosed. The bits in the speech data packet (10) are sorted in a predetermined error sensitivity characteristic, order, level or degree of importance. Only those bits in the packet (10) which are considered to be most error sensitive are protected by redundant transmission. A partial set of redundant bits of the previously transmitted packets (10) are included with the data bit for current packet (10). The redundant bits are used at the receiver side to reconstruct damaged packets. By using only the most sensitive bits for redundancy, the additional required bandwith may be limited.

Abstract translation: 公开了一种用于语音数据分组的部分冗余编码的方法和装置。 语音数据分组（10）中的比特以预定的误差灵敏度特征，顺序，级别或重要程度进行排序。 被认为是最大误差的数据包（10）中的那些位只能被冗余传输保护。 先前发送的分组（10）的部分冗余比特包括在当前分组（10）的数据比特中。 在接收端使用冗余比特来重构损坏的数据包。 通过仅使用最灵敏的位进行冗余，附加的所需带宽可能受到限制。

公开(公告)号：WO0016315A3

公开(公告)日：2000-05-25

申请号：PCT/SE9901433

申请日：1999-08-24

Applicant: ERICSSON TELEFON AB L M

Inventor： EKUDDEN ERIK ,  HAGEN ROAR

IPC: H03M7/30 ,  G10L19/083 ,  H03M7/36 ,  G10L19/12

CPC classification number: G10L19/083

Abstract: A linear predictive analysis-by-synthesis encoder includes a search algorithm block (50) and a vector quantizer (58) for vector quantizing optimal gains from a plurality of subframes in a frame. The internal encoder states are updated using (50, 52, 54, 56) the vector quantized gains.

Abstract translation: 合成线性预测分析编码器包括用于向量量化来自帧中多个子帧的最优增益的搜索算法块（50）和矢量量化器（58）。 使用（50,52,54,56）矢量量化增益来更新内部编码器状态。

公开(公告)号：WO9959282A3

公开(公告)日：2000-02-03

申请号：PCT/SE9900689

申请日：1999-04-28

Applicant: ERICSSON TELEFON AB L M

Inventor： EKUDDEN ERIK ,  SVENSSON TOMAS ,  SUNDQVIST JIM

IPC: G10L19/12 ,  G10L19/00 ,  H04L1/00 ,  H04L12/56 ,  H04L12/64 ,  H04L1/22

CPC classification number: H04L12/6418 ,  H04L2012/6481

Abstract: A decoder improves delayed packet concealment in a packet network by using two decoder sections. A first decoder section (30) bases its decoding during the concealment phase on erroneous filter states and a set of speech parameters, whereas a second decoder section bases its decoding on saved (36) and updated filter states and the same speech parameters. The outputs of the two decoder sections are thereafter combined (34) to form the final speech signal. This decoding strategy produces a speech signal with smooth transitions from delayed to non-delayed packets and uses information from the most recent packets for speech generation.

Abstract translation: 解码器通过使用两个解码器部分来改进分组网络中的延迟分组隐藏。 第一解码器部分（30）在隐藏阶段的解码基于错误滤波器状态和一组语音参数，而第二解码器部分基于其保存的（36）和更新的滤波器状态和相同语音参数的解码。 此后，组合两个解码器部分的输出（34）以形成最终语音信号。 该解码策略产生具有从延迟到非延迟分组的平滑过渡的语音信号，并且使用来自最新分组的信息用于语音产生。

公开(公告)号：CA2349944C

公开(公告)日：2010-01-12

申请号：CA2349944

申请日：1999-11-08

Applicant: ERICSSON TELEFON AB L M

Inventor： EKUDDEN ERIK ,  HAGEN ROAR ,  JOHANSSON INGEMAR

IPC: G10L19/012 ,  G10L19/14 ,  G10L19/00 ,  G10L19/04 ,  H04B14/04 ,  H04M1/00 ,  H04M1/253 ,  H04M1/725

Abstract: The quality of comfort noise generated by a speech decoder (93) during non- speech periods is improved by modifying (30,75) comfort noise parameter values (33) normally used to generate the comfort noise. The comfort noise parameter values are modified in response to variability information (43) associated with a background noise parameter. The modified comfort noise parameter values (35) are then used to generate t he comfort noise.

公开(公告)号：DE60132321T2

公开(公告)日：2008-12-11

申请号：DE60132321

申请日：2001-04-19

Applicant: ERICSSON TELEFON AB L M

Inventor： EKUDDEN ERIK ,  ERIKSSON ANDERS

IPC: G10L21/02 ,  G10L19/00

Abstract: The present invention advantageously provides a manner by which to further suppress noise superimposed upon an information signal without increasing distortion to the signal, e.g., speech. By distributing the noise suppression, the quality of the information signal provided to a listener is improved. In one embodiment, a first noise suppressor is employed at the transmitter to suppress noise superimposed upon an information signal prior to its transmission by the transmitter, and a second noise suppressor is employed at the receiver to suppress the noise component of a communication signal received at the receiver.

公开(公告)号：DE69930929T2

公开(公告)日：2006-12-07

申请号：DE69930929

申请日：1999-04-28

Applicant: ERICSSON TELEFON AB L M

Inventor： EKUDDEN ERIK ,  SVENSSON TOMAS ,  SUNDQVIST JIM

IPC: G10L19/12 ,  H04L12/56 ,  G10L19/00 ,  H04L1/00 ,  H04L1/22 ,  H04L12/64

Abstract: A decoder improves delayed packet concealment in a packet network by using two decoder sections. A first decoder section bases its decoding during the concealment phase on erroneous filter states and a set of speech parameters, whereas a second decoder section bases its decoding on saved and updated filter states and the same speech parameters. The outputs of the two decoder sections are thereafter combined to form the final speech signal. This decoding strategy produces a speech signal with smooth transitions from delayed to non-delayed packets and uses information from the most recent packets for speech generation.

公开(公告)号：DE69922388T2

公开(公告)日：2005-12-22

申请号：DE69922388

申请日：1999-08-24

Applicant: ERICSSON TELEFON AB L M

Inventor： EKUDDEN ERIK ,  HAGEN ROAR

IPC: H03M7/30 ,  G10L19/083 ,  H03M7/36 ,  G10L19/14

Abstract: A linear predictive analysis-by-synthesis encoder includes a search algorithm block (50) and a vector quantizer (58) for vector quantizing optimal gains from a plurality of subframes in a frame. The internal encoder states are updated (50, 52, 54, 56) using the vector quantized gains.