-
公开(公告)号：US12198057B2
公开(公告)日：2025-01-14
申请号：US17102748
申请日：2020-11-24
Applicant: QUALCOMM Incorporated
Inventor： Fatemeh Saki , Yinyi Guo , Erik Visser
Abstract: A device includes one or more processors configured to receive sensor data from one or more sensor devices. The one or more processors are also configured to determine a context of the device based on the sensor data. The one or more processors are further configured to select a model based on the context. The one or more processors are also configured to process an input signal using the model to generate a context-specific output.
-
公开(公告)号：US11240058B2
公开(公告)日：2022-02-01
申请号：US16370836
申请日：2019-03-29
Applicant: QUALCOMM Incorporated
Inventor： Ravi Choudhary , Yinyi Guo , Fatemeh Saki , Erik Visser
Abstract: A device to provide information to a visual interface that is mountable to a vehicle dashboard includes a memory configured to store device information indicative of controllable devices of a building and occupant data indicative of one or more occupants of the building. The device includes a processor configured to receive, in real-time, status information associated with the one or more occupants of the building. The status information includes at least one of dynamic location information or dynamic activity information. The processor is configured to generate an output to provide, at the visual interface device, a visual representation of at least a portion of the building and the status information associated with the one or more occupants. The processor is also configured to generate an instruction to adjust an operation of one or more devices of the controllable devices based on user input.
-
公开(公告)号：US10540979B2
公开(公告)日：2020-01-21
申请号：US14688974
申请日：2015-04-16
Applicant: QUALCOMM Incorporated
Inventor： Erik Visser , Lae-Hoon Kim , Minho Jin , Yinyi Guo
Abstract: A device includes a memory, a receiver, a processor, and a display. The memory is configured to store a speaker model. The receiver is configured to receive an input audio signal. The processor is configured to determine a first confidence level associated with a first portion of the input audio signal based on the speaker model. The processor is also configured to determine a second confidence level associated with a second portion of the input audio signal based on the speaker model. The display is configured to present a graphical user interface associated with the first confidence level or associated with the second confidence level.
-
公开(公告)号：US20190320281A1
公开(公告)日：2019-10-17
申请号：US15951907
申请日：2018-04-12
Applicant: QUALCOMM Incorporated
Inventor： Yinyi Guo , Lae-Hoon Kim , Dongmei Wang , Erik Visser
Abstract: An apparatus includes a processor configured to receive one or more media signals associated with a scene. The processor is also configured to identify a spatial location in the scene for each source of the one or more media signals. The processor is further configured to identify audio content for each media signal of the one or more media signals. The processor is also configured to determine one or more candidate spatial locations in the scene based on the identified spatial locations. The processor is further configured to generate audio to playback as virtual sounds that originate from the one or more candidate spatial locations.
-
5.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
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。
-
Patent Agency Ranking
公开(公告)号：US11094316B2
公开(公告)日：2021-08-17
申请号：US15972011
申请日：2018-05-04
Applicant: QUALCOMM Incorporated
Inventor： Erik Visser , Fatemeh Saki , Yinyi Guo , Sunkuk Moon , Lae-Hoon Kim , Ravi Choudhary
Abstract: A device includes a memory configured to store category labels associated with categories of a natural language processing library. A processor is configured to analyze input audio data to generate a text string and to perform natural language processing on at least the text string to generate an output text string including an action associated with a first device, a speaker, a location, or a combination thereof. The processor is configured to compare the input audio data to audio data of the categories to determine whether the input audio data matches any of the categories and, in response to determining that the input audio data does not match any of the categories: create a new category label, associate the new category label with at least a portion of the output text string, update the categories with the new category label, and generate a notification indicating the new category label.
-
公开(公告)号：US20190341026A1
公开(公告)日：2019-11-07
申请号：US15972011
申请日：2018-05-04
Applicant: QUALCOMM Incorporated
Inventor： Erik Visser , Fatemeh Saki , Yinyi Guo , Sunkuk Moon , Lae-Hoon Kim , Ravi Choudhary
Abstract: A device includes a memory configured to store category labels associated with categories of a natural language processing library. A processor is configured to analyze input audio data to generate a text string and to perform natural language processing on at least the text string to generate an output text string including an action associated with a first device, a speaker, a location, or a combination thereof. The processor is configured to compare the input audio data to audio data of the categories to determine whether the input audio data matches any of the categories and, in response to determining that the input audio data does not match any of the categories: create a new category label, associate the new category label with at least a portion of the output text string, update the categories with the new category label, and generate a notification indicating the new category label.
-
公开(公告)号：US20180233127A1
公开(公告)日：2018-08-16
申请号：US15430791
申请日：2017-02-13
Applicant: QUALCOMM Incorporated
Inventor： ERIK VISSER , Shuhua Zhang , Lae-Hoon Kim , Yinyi Guo , Sunkuk Moon
IPC: G10L13/027 , G10L13/047 , G10L25/78 , G10L25/21 , G10L25/63 , G10L25/90 , G10L15/26
CPC classification number: G10L15/26 , G10L13/047 , G10L21/00 , G10L21/003 , G10L25/48
Abstract: In a particular aspect, an apparatus includes an audio sensor configured to receive an input audio signal. The apparatus also includes speech generative circuitry configured to generate a synthesized audio signal based at least partly on automatic speech recognition (ASR) data associated with the input audio signal and based on one or more parameters indicative of state information associated with the input audio signal.
-
公开(公告)号：US09666183B2
公开(公告)日：2017-05-30
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: G10L15/00 , G10L15/16 , G10L25/51 , G10L25/30 , G10L13/00 , G10L15/02 , G10L15/06 , G10L21/0208 , G10L21/0272
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.
-
公开(公告)号：US09536537B2
公开(公告)日：2017-01-03
申请号：US14634637
申请日：2015-02-27
Applicant: QUALCOMM Incorporated
Inventor： Yinyi Guo , Shuhua Zhang , Erik Visser , Lae-Hoon Kim , Sanghyun Chi
IPC: G10L21/0208 , G10L21/034 , H03G5/00 , H03G5/16 , G10L21/0232
CPC classification number: G10L21/0208 , G10L21/0232 , G10L21/034 , H03G5/005 , H03G5/165
Abstract: A method for speech restoration by an electronic device is described. The method includes obtaining a noisy speech signal. The method also includes suppressing noise in the noisy speech signal to produce a noise-suppressed speech signal. The noise-suppressed speech signal has a bandwidth that includes at least three subbands. The method further includes iteratively restoring each of the at least three subbands. Each of the at least three subbands is restored based on all previously restored subbands of the at least three subbands.
Abstract translation: 描述了一种通过电子设备进行语音恢复的方法。 该方法包括获得噪声语音信号。 该方法还包括抑制噪声语音信号中的噪声以产生噪声抑制语音信号。 噪声抑制语音信号具有包括至少三个子带的带宽。 该方法还包括迭代地恢复至少三个子带中的每一个。 所述至少三个子带中的每一个基于所述先前恢复的所述至少三个子带的子带被恢复。
-
-
-
-
-
-
-
-
-
Search Results
44
records
(took 0.02 seconds)
