公开(公告)号：US20160005406A1

公开(公告)日：2016-01-07

申请号：US14767279

申请日：2014-01-22

Applicant: Dolby Laboratories Licensing Corporation

Inventor： Kuan-Chieh YEN ,  Vinay MELKOTE ,  Matthew FELLERS ,  Grant A. DAVIDSON

IPC: G10L19/008 ,  H04S3/00

CPC classification number: G10L19/008 ,  H04S3/008 ,  H04S5/00 ,  H04S2400/03 ,  H04S2420/03 ,  H04S2420/07

Abstract: Audio characteristics of audio data corresponding to a plurality of audio channels may be determined. The audio characteristics may include spatial parameter data. Decorrelation filtering processes for the audio data may be based, at least in part, on the audio characteristics. The decorrelation filtering processes may cause a specific inter-decorrelation signal coherence (“IDC”) between channel-specific decorrelation signals for at least one pair of channels. The channel-specific decorrelation signals may be received and/or determined. Inter-channel coherence (“ICC”) between a plurality of audio channel pairs may be controlled. Controlling ICC may involve at receiving an ICC value and/or determining an ICC value based, at least partially, on the spatial parameter data. A set of IDC values may be based, at least partially, on the set of ICC values. A set of channel-specific decorrelation signals, corresponding with the set of IDC values, may be synthesized by performing operations on the filtered audio data.

Abstract translation: 可以确定与多个音频通道对应的音频数据的音频特性。 音频特征可以包括空间参数数据。 音频数据的解相关滤波处理可以至少部分地基于音频特性。 去相关滤波处理可以在至少一对信道之间引起信道特定解相关信号之间的特定解相关信号相干性（“IDC”）。 信道特定的去相关信号可以被接收和/或确定。 可以控制多个音频通道对之间的通道间相干（“ICC”）。 控制ICC可以涉及至少部分地基于空间参数数据接收ICC值和/或确定ICC值。 一组IDC值可以至少部分地基于ICC值集合。 可以通过对经滤波的音频数据执行操作来合成与集合的IDC值相对应的一组通道特定的解相关信号。

公开(公告)号：US20160005405A1

公开(公告)日：2016-01-07

申请号：US14766957

申请日：2014-01-22

Applicant: DOLBY LABORATORIES LICENSING CORPORATION

Inventor： Kuan-Chieh YEN ,  Vinay MELKOTE ,  Grant A. DAVIDSON

IPC: G10L19/008 ,  H04S5/00

CPC classification number: G10L19/008 ,  G10H7/008 ,  G10H2250/225 ,  H04S5/00 ,  H04S2400/03 ,  H04S2420/07

Abstract: Some audio processing methods may involve receiving audio data corresponding to a plurality of audio channels and determining audio characteristics of the audio data, which may include transient information. An amount of decorrelation for the audio data may be based, at least in part, on the audio characteristics. If a definite transient event is determined, a decorrelation process may be temporarily halted or slowed. Determining transient information may involve evaluating the likelihood and/or the severity of a transient event. In some implementations, determining transient information may involve evaluating a temporal power variation in the audio data. Explicit transient information may or may not be received with the audio data, depending on the implementation. Explicit transient information may include a transient control value corresponding to a definite transient event, a definite non-transient event or an intermediate transient control value.

Abstract translation: 一些音频处理方法可以包括接收对应于多个音频信道的音频数据并确定音频数据的音频特性，其可以包括瞬时信息。 用于音频数据的去相关的量可以至少部分地基于音频特性。 如果确定了一个确定的瞬时事件，则解相关过程可能暂时停止或减慢。 确定瞬时信息可能涉及评估瞬态事件的可能性和/或严重性。 在一些实现中，确定瞬时信息可以包括评估音频数据中的时间功率变化。 取决于实现情况，显式瞬时信息可能会或可能不会与音频数据一起接收。 显式瞬态信息可以包括对应于确定的瞬态事件，确定的非瞬态事件或中间瞬态控制值的瞬态控制值。

公开(公告)号：US20240079019A1

公开(公告)日：2024-03-07

申请号：US18507824

申请日：2023-11-13

Applicant: Dolby Laboratories Licensing Corporation

Inventor： Roy M. FEJGIN ,  Grant A. DAVIDSON ,  Chih-Wei WU ,  Vivek KUMAR

IPC: G10L19/022 ,  G06F3/16 ,  G06N3/048 ,  G06N3/084

CPC classification number: G10L19/022 ,  G06F3/16 ,  G06N3/048 ,  G06N3/084

Abstract: Computer-implemented methods for training a neural network, as well as for implementing audio encoders and decoders via trained neural networks, are provided. The neural network may receive an input audio signal, generate an encoded audio signal and decode the encoded audio signal. A loss function generating module may receive the decoded audio signal and a ground truth audio signal, and may generate a loss function value corresponding to the decoded audio signal. Generating the loss function value may involve applying a psychoacoustic model. The neural network may be trained based on the loss function value. The training may involve updating at least one weight of the neural network.

公开(公告)号：US20230386486A1

公开(公告)日：2023-11-30

申请号：US18248294

申请日：2021-10-15

Applicant: Dolby Laboratories Licensing Corporation

Inventor： Cong ZHOU ,  Grant A. DAVIDSON ,  Mark S. VINTON

IPC: G10L19/022 ,  G10L25/30 ,  G10L19/032 ,  G10L19/04

CPC classification number: G10L19/022 ,  G10L19/04 ,  G10L19/032 ,  G10L25/30

Abstract: The present invention relates to a method for predicting transform coefficients representing frequency content of an adaptive block length media signal, by receiving a frame and receiving block length information indicating a number of quantized transform coefficients for each block in the frame, the number of quantized transform coefficients being one of a first or second number, wherein the first number is greater than the second number, determining a first block has the second number of quantized transform coefficients, converting the first block into a converted block having the first number of quantized transform coefficients, conditioning a main neural network trained to predict at least one output variable given at least one conditioning variable, the at least one conditioning variable being based on information regarding the converted block and block length information for the first block, providing at least one predicted transform coefficients from an output stage of the main neural network.

公开(公告)号：US20160005413A1

公开(公告)日：2016-01-07

申请号：US14767565

申请日：2014-01-22

Applicant: DOLBY LABORATORIES LICENSING CORPORATION

Inventor： Matthew FELLERS ,  Vinay MELKOTE ,  Kuan-Chieh YEN ,  Grant A. DAVIDSON ,  Mark F. DAVIS

IPC: G10L19/02 ,  G10L19/012 ,  G10L25/18 ,  G10L19/008 ,  G10L25/06

CPC classification number: G10L19/008 ,  G10L19/012 ,  G10L19/0204 ,  G10L25/06 ,  G10L25/18

Abstract: Received audio data may include a first set of frequency coefficients and a second set of frequency coefficients. Spatial parameters for at least part of the second set of frequency coefficients may be estimated, based at least in part on the first set of frequency coefficients. The estimated spatial parameters may be applied to the second set of frequency coefficients to generate a modified second set of frequency coefficients. The first set of frequency coefficients may correspond to a first frequency range (for example, an individual channel frequency range) and the second set of frequency coefficients may correspond to a second frequency range (for example, a coupled channel frequency range). Combined frequency coefficients of a composite coupling channel may be based on frequency coefficients of two or more channels. Cross-correlation coefficients, between frequency coefficients of a first channel and the combined frequency coefficients, may be computed.

Abstract translation: 接收的音频数据可以包括第一组频率系数和第二组频率系数。 可以至少部分地基于第一组频率系数来估计第二组频率系数的至少一部分的空间参数。 估计的空间参数可以应用于第二组频率系数，以产生经修改的第二组频率系数。 第一组频率系数可以对应于第一频率范围（例如，单个信道频率范围），并且第二组频率系数可以对应于第二频率范围（例如，耦合的信道频率范围）。 复合耦合信道的组合频率系数可以基于两个或更多个信道的频率系数。 可以计算第一通道的频率系数与组合频率系数之间的互相关系数。

公开(公告)号：US20150371646A1

公开(公告)日：2015-12-24

申请号：US14767867

申请日：2014-01-22

Applicant: Dolby Laboratories Licensing Corporation

Inventor： Vinay MELKOTE ,  Kuan-Chieh YEN ,  Matthew FELLERS ,  Grant A. DAVIDSON ,  Vivek KUMAR

IPC: G10L19/008 ,  G10L25/06 ,  G10L25/09 ,  H04S3/00

CPC classification number: G10L19/008 ,  G10H2250/061 ,  G10H2250/225 ,  G10H2250/575 ,  G10L25/06 ,  G10L25/09 ,  H04S3/008 ,  H04S5/00 ,  H04S2400/01 ,  H04S2400/03 ,  H04S2420/03 ,  H04S2420/07

Abstract: Decorrelation filter parameters for audio data may be based, at least in part, on audio characteristics such as tonality information and/or transient information. Determining the audio characteristics may involve receiving explicit audio characteristics with the audio data and/or determining audio characteristics based on one or more attributes of the audio data. The decorrelation filter parameters may include dithering parameters and/or randomly selected pole locations for at least one pole of an all-pass filter. The dithering parameters and/or pole locations may involve a maximum stride value for pole movement. In some examples, the maximum stride value may be substantially zero for highly tonal signals of the audio data. The dithering parameters and/or pole locations may be bounded by constraint areas within which pole movements are constrained. The constraint areas may or may not be fixed. In some implementations, different channels of the audio data may share the same constraint areas.

Abstract translation: 音频数据的解相关滤波器参数可以至少部分地基于诸如音调信息和/或瞬态信息的音频特性。 确定音频特征可以包括基于音频数据的一个或多个属性来接收具有音频数据的显式音频特征和/或确定音频特性。 去相关滤波器参数可以包括全通滤波器的至少一个极点的抖动参数和/或随机选择的极点位置。 抖动参数和/或极点位置可能涉及极运动的最大步幅值。 在一些示例中，对于音频数据的高音调信号，最大步幅值可以基本为零。 抖动参数和/或极点位置可以限制极限运动受限制的约束区域。 约束区域可以是或可以不是固定的。 在一些实现中，音频数据的不同频道可以共享相同的约束区域。

公开(公告)号：US20140324441A1

公开(公告)日：2014-10-30

申请号：US14325130

申请日：2014-07-07

Applicant: DOLBY LABORATORIES LICENSING CORPORATION ,  DOLBY INTERNATIONAL AB

Inventor： Arijit BISWAS ,  Vinay MELKOTE ,  Michael SCHUG ,  Grant A. DAVIDSON ,  Mark S. VINTON

IPC: G10L19/028 ,  G10L19/26

CPC classification number: G10L19/028 ,  G10L19/0204 ,  G10L19/032 ,  G10L19/265

Abstract: A method for determining mantissa bit allocation of audio data values of frequency domain audio data to be encoded. The allocation method includes a step of determining masking values for the audio data values, including by performing adaptive low frequency compensation on the audio data of each frequency band of a set of low frequency bands of the audio data. The adaptive low frequency compensation includes steps of: performing tonality detection on the audio data to generate compensation control data indicative of whether each frequency band in the set of low frequency bands has prominent tonal content; and performing low frequency compensation on the audio data in each frequency band in the set of low frequency bands having prominent tonal content as indicated by the compensation control data, but not performing low frequency compensation on the audio data in any other frequency band in the set of low frequency bands.

Abstract translation: 一种用于确定要编码的频域音频数据的音频数据值的尾数位分配的方法。 分配方法包括通过对音频数据的一组低频带的每个频带的音频数据执行自适应低频补偿来确定音频数据值的屏蔽值的步骤。 所述自适应低频补偿包括以下步骤：对所述音频数据执行音调检测，以产生指示所述一组低频带中的每个频带是否具有突出的音调内容的补偿控制数据; 对由该补偿控制数据所表示的具有突出色调内容的低频带组中的每个频带中的音频数据执行低频补偿，而不对该组中的任何其它频带中的音频数据执行低频补偿 的低频带。

公开(公告)号：US20250080943A1

公开(公告)日：2025-03-06

申请号：US18829074

申请日：2024-09-09

Applicant: DOLBY LABORATORIES LICENSING CORPORATION

Inventor： Kuan-Chieh YEN ,  Dirk Jeroen BREEBAART ,  Grant A. DAVIDSON ,  Rhonda WILSON ,  David M. COOPER ,  Zhiwei SHUANG

IPC: H04S7/00 ,  G10L19/008 ,  H04S3/00

Abstract: In some embodiments, virtualization methods for generating a binaural signal in response to channels of a multi-channel audio signal, which apply a binaural room impulse response (BRIR) to each channel including by using at least one feedback delay network (FDN) to apply a common late reverberation to a downmix of the channels. In some embodiments, input signal channels are processed in a first processing path to apply to each channel a direct response and early reflection portion of a single-channel BRIR for the channel, and the downmix of the channels is processed in a second processing path including at least one FDN which applies the common late reverberation. Typically, the common late reverberation emulates collective macro attributes of late reverberation portions of at least some of the single-channel BRIRs. Other aspects are headphone virtualizers configured to perform any embodiment of the method.

公开(公告)号：US20240430637A1

公开(公告)日：2024-12-26

申请号：US18759221

申请日：2024-06-28

Applicant: DOLBY LABORATORIES LICENSING CORPORATION

Inventor： Grant A. DAVIDSON ,  Kuan-Chieh YEN ,  Dirk Jeroen BREEBAART

IPC: H04S7/00

Abstract: Methods and systems for designing binaural room impulse responses (BRIRs) for use in headphone virtualizers, and methods and systems for generating a binaural signal in response to a set of channels of a multi-channel audio signal, including by applying a BRIR to each channel of the set, thereby generating filtered signals, and combining the filtered signals to generate the binaural signal, where each BRIR has been designed in accordance with an embodiment of the design method. Other aspects are audio processing units configured to perform any embodiment of the inventive method. In accordance with some embodiments, BRIR design is formulated as a numerical optimization problem based on a simulation model (which generates candidate BRIRs) and at least one objective function (which evaluates each candidate BRIR), and includes identification of a best one of the candidate BRIRs as indicated by performance metrics determined for the candidate BRIRs by each objective function.

公开(公告)号：US20230395086A1

公开(公告)日：2023-12-07

申请号：US18031790

申请日：2021-10-14

Applicant: DOLBY LABORATORIES LICENSING CORPORATION

Inventor： Mark S. VINTON ,  Cong ZHOU ,  Roy M. FEJGIN ,  Grant A. DAVIDSON

IPC: G10L19/032 ,  G10L19/06

CPC classification number: G10L19/032 ,  G10L19/06

Abstract: Described herein is a method of processing an audio signal using a neural network or using a first and a second neural network. Described is further a method of training said neural network or of jointly training a set of said first and said second neural network. Moreover, described is a method of obtaining and transmitting a latent feature space representation of a perceptual domain audio signal using a neural network and a method of obtaining an audio signal from a latent feature space representation of a perceptual domain audio signal using a neural network. Described are also respective apparatuses and computer program products.