知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

158 records (took 0.028 seconds)

    DEEP NEURAL NET BASED FILTER PREDICTION FOR AUDIO EVENT CLASSIFICATION AND EXTRACTION
    16.
    发明申请
    DEEP NEURAL NET BASED FILTER PREDICTION FOR AUDIO EVENT CLASSIFICATION AND EXTRACTION 有权
    深度基于神经网络的滤波器预测音频事件分类和提取

    公开(公告)号：US20160284346A1

    公开(公告)日：2016-09-29

    申请号：US14671850

    申请日：2015-03-27

    Applicant: QUALCOMM Incorporated

    Inventor： Erik Visser ,  Yinyi Guo ,  Lae-Hoon Kim ,  Raghuveer Peri ,  Shuhua Zhang

    IPC: G10L15/16 ,  G10L15/06 ,  G10L15/02

    CPC classification number: G10L15/16 ,  G10L15/02 ,  G10L15/063 ,  G10L21/0208 ,  G10L21/0272 ,  G10L25/30 ,  G10L25/51

    Abstract: Disclosed is a feature extraction and classification methodology wherein audio data is gathered in a target environment under varying conditions. From this collected data, corresponding features are extracted, labeled with appropriate filters (e.g., audio event descriptions), and used for training deep neural networks (DNNs) to extract underlying target audio events from unlabeled training data. Once trained, these DNNs are used to predict underlying events in noisy audio to extract therefrom features that enable the separation of the underlying audio events from the noisy components thereof.

    Abstract translation: 公开了一种特征提取和分类方法，其中音频数据在不同条件下收集在目标环境中。 从该收集的数据中，提取相应的特征，用适当的滤波器（例如，音频事件描述）标记，并用于训练深层神经网络（DNN）以从未标记的训练数据中提取潜在的目标音频事件。 一旦被训练，这些DNN用于预测嘈杂音频中的底层事件以从其中提取能够将底层音频事件与其噪声分量分开的特征。

    METHOD, SYSTEM AND ARTICLE OF MANUFACTURE FOR PROCESSING SPATIAL AUDIO
    17.
    发明申请
    METHOD, SYSTEM AND ARTICLE OF MANUFACTURE FOR PROCESSING SPATIAL AUDIO 有权
    用于处理空间音频的制造方法，系统和制品

    公开(公告)号：US20160198282A1

    公开(公告)日：2016-07-07

    申请号：US14807760

    申请日：2015-07-23

    Applicant: QUALCOMM Incorporated

    Inventor： Lae-Hoon Kim ,  Raghuveer Peri ,  Erik Visser

    IPC: H04S7/00 ,  H04S3/00 ,  G06F3/16

    CPC classification number: H04S7/30 ,  G06F3/165 ,  H04S3/002 ,  H04S7/301 ,  H04S2400/01 ,  H04S2400/11 ,  H04S2400/15

    Abstract: Techniques for processing directionally-encoded audio to account for spatial characteristics of a listener playback environment are disclosed. The directionally-encoded audio data includes spatial information indicative of one or more directions of sound sources in an audio scene. The audio data is modified based on input data identifying the spatial characteristics of the playback environment. The spatial characteristics may correspond to actual loudspeaker locations in the playback environment. The directionally-encoded audio may also be processed to permit focusing/defocusing on sound sources or particular directions in an audio scene. The disclosed techniques may allow a recorded audio scene to be more accurately reproduced at playback time, regardless of the output loudspeaker setup. Another advantage is that a user may dynamically configure audio data so that it better conforms to the user's particular loudspeaker layouts and/or the user's desired focus on particular subjects or areas in an audio scene.

    Abstract translation: 公开了用于处理定向编码的音频以考虑收听者回放环境的空间特征的技术。 定向编码的音频数据包括指示音频场景中的声源的一个或多个方向的空间信息。 基于识别回放环境的空间特性的输入数据来修改音频数据。 空间特征可以对应于播放环境中的实际扬声器位置。 定向编码的音频也可以被处理以允许对音频场景中的声源或特定方向进行聚焦/散焦。 所公开的技术可以允许在播放时间更准确地再现记录的音频场景，而与输出的扬声器设置无关。 另一个优点是用户可以动态地配置音频数据，使得其更好地符合用户的特定扬声器布局和/或用户对音频场景中的特定主体或区域的期望焦点。

    Listen to people you recognize
    19.
    发明授权
    Listen to people you recognize 有权
    听你认识的人

    公开(公告)号：US09282399B2

    公开(公告)日：2016-03-08

    申请号：US14191321

    申请日：2014-02-26

    Applicant: QUALCOMM Incorporated

    Inventor： Lae-Hoon Kim ,  Phuong Lam Ton ,  Erik Visser ,  Jeremy P. Toman ,  Francis Bernard MacDougall

    IPC: H04R1/32 ,  G10L15/25 ,  G06K9/00 ,  G10L21/02 ,  H04N7/14 ,  G10L21/0216

    CPC classification number: H04R3/005 ,  G06K9/00221 ,  G06K9/00315 ,  G10L15/25 ,  G10L21/02 ,  G10L21/0208 ,  G10L2021/02166 ,  H04N7/147 ,  H04R1/326 ,  H04R1/406

    Abstract: Systems, devices, and methods are described for recognizing and focusing on at least one source of an audio communication as part of a communication including a video image and an audio communication derived from two or more microphones when a relative position between the microphones is known. In certain embodiments, linked audio and video focus areas providing location information for one or more sound sources may each be associated with different user inputs, and an input to adjust a focus in either the audio or video domain may automatically adjust the focus in the another domain.

    Abstract translation: 描述了系统，设备和方法，用于当麦克风之间的相对位置是已知的时，用于识别和聚焦至少一个音频通信源，作为包括从两个或更多个麦克风导出的视频图像和音频通信的通信的一部分。 在某些实施例中，提供一个或多个声源的位置信息的链接的音频和视频焦点区域可以各自与不同的用户输入相关联，并且用于调整音频或视频域中的焦点的输入可以自动调整另一个 域。

    Content based noise suppression
    20.
    发明授权
    Content based noise suppression 有权
    基于内容的噪声抑制

    公开(公告)号：US09275625B2

    公开(公告)日：2016-03-01

    申请号：US13787605

    申请日：2013-03-06

    Applicant: QUALCOMM Incorporated

    Inventor： Lae-Hoon Kim ,  Asif Iqbal Mohammad ,  Erik Visser ,  Jongwon Shin

    IPC: A61F11/06 ,  G10K11/16 ,  H04M9/08 ,  G10L21/0216

    CPC classification number: G10K11/16 ,  G10L21/0216 ,  H04M9/082

    Abstract: Apparatus and methods for audio noise attenuation are disclosed. An audio signal analyzer can determine whether an input audio signal received from a microphone device includes a noise signal having identifiable content. If there is a noise signal having identifiable content, a content source is accessed to obtain a copy of the noise signal. An audio canceller can generate a processed audio signal, having an attenuated noise signal, based on comparing the copy of the noise signal to the input audio signal. Additionally or alternatively, data may be communicated on a communication channel to a separate media device to receive at least a portion of the copy of the noise signal from the separate media device, or to receive content-identification data corresponding to the content source.

    Abstract translation: 公开了用于音频噪声衰减的装置和方法。 音频信号分析器可以确定从麦克风装置接收的输入音频信号是否包括具有可识别内容的噪声信号。 如果存在具有可识别内容的噪声信号，则访问内容源以获得噪声信号的副本。 基于将噪声信号的复制与输入音频信号进行比较，音频消除器可以生成具有衰减噪声信号的已处理音频信号。 另外或替代地，数据可以在通信信道上传送到单独的媒体设备，以从单独的媒体设备接收噪声信号的副本的至少一部分，或者接收对应于内容源的内容标识数据。

Patent Agency Ranking