Patents by Inventor Yen-Hsun Chu

Yen-Hsun Chu has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Patent number: 11989940
    Abstract: A movie detection system and a movie detection method are provided. The method includes: capturing a current image of an input video source; performing a pre-processing process and an image scaling on the current image, cutting and retaining a top and a bottom of the current image, and joining the top and bottom to generate a joined current image; inputting the joined current image into a trained machine learning model to classify the joined current image as a movie image or a non-movie image. The trained machine learning model is generated by performing a training process on a machine learning model, and the training process is performed based on a plurality of joined training images. The joined training images are generated by performing the pre-processing process on a plurality of movie images with black bars and a plurality of non-movie images without black bars.
    Type: Grant
    Filed: December 15, 2021
    Date of Patent: May 21, 2024
    Assignee: REALTEK SEMICONDUCTOR CORP.
    Inventor: Yen-Hsun Chu
  • Publication number: 20240046926
    Abstract: A television includes a remote control, a receiving element, a speaker, a speech analysis model, and a processor. The processor analyzes video sound to get a repeated audio section after receiving a volume adjustment command from the remote control. Then, the speaker outputs the repeated audio. So that, according to user needs, the television adjusts the video sound before outputting.
    Type: Application
    Filed: October 24, 2022
    Publication date: February 8, 2024
    Applicant: REALTEK SEMICONDUCTOR CORP.
    Inventor: Yen-Hsun Chu
  • Publication number: 20240046949
    Abstract: An audio real-time processing system, an audio real-time processing program product and method for training speech analysis model are provided. The speech analysis model is firstly trained to obtain, from an original audio, mask information which is used to mask the original audio to get a target audio. The system obtains a plurality of analyzed audio according to the target audio and the original audio, obtains repeated audio section according to the plurality of the analyzed and output the repeated audio section.
    Type: Application
    Filed: October 24, 2022
    Publication date: February 8, 2024
    Applicant: REALTEK SEMICONDUCTOR CORP.
    Inventor: Yen-Hsun CHU
  • Publication number: 20230154189
    Abstract: A video classification system and method and a neural network training system are provided. The video classification system captures multiple sampled images, encodes first and second images of all sampled images into feature matrices of the sampled images using a convolutional neural network module, and obtains the classification of the videos based on a recurrent neural network module and the feature matrices. The neural network training system captures multiple first sampled images and uses first training samples of all the first sampled images to train the convolutional neural network module and a classification module to obtain multiple parameter values of the convolutional neural network module. The neural network training system trains the recurrent neural network module based on the parameter values and multiple second sampled images.
    Type: Application
    Filed: May 2, 2022
    Publication date: May 18, 2023
    Applicant: REALTEK SEMICONDUCTOR CORP.
    Inventors: Ling SUN, Yen-Hsun CHU
  • Patent number: 11495242
    Abstract: The audio processing method may include: utilizing a first convolution layer circuit and a second convolution layer circuit to perform convolution processing on an audio mixture signal, respectively, in order to generate an input feature map; performing separation processing corresponding to a predetermined neural network model on the input feature map to obtain a first output feature map and a second output feature map respectively corresponding to a first type and a second type; and utilizing a first deconvolution layer circuit and a second deconvolution layer circuit to perform deconvolution processing on different portions of the first output feature map, respectively, and utilizing the first deconvolution layer circuit and the second deconvolution layer circuit to perform deconvolution processing on different portions of the second output feature map, respectively, in order to generate corresponding processing results, for generating multiple audio output signals.
    Type: Grant
    Filed: May 25, 2021
    Date of Patent: November 8, 2022
    Assignee: Realtek Semiconductor Corp.
    Inventors: Tai-Shih Chi, Ya-Ching Lai, Yih-Liang Shen, Yen-Hsun Chu, Chia-Chun Cheng
  • Publication number: 20220292824
    Abstract: A movie detection system and a movie detection method are provided. The method includes: capturing a current image of an input video source; performing a pre-processing process and an image scaling on the current image, cutting and retaining a top and a bottom of the current image, and joining the top and bottom to generate a joined current image; inputting the joined current image into a trained machine learning model to classify the joined current image as a movie image or a non-movie image. The trained machine learning model is generated by performing a training process on a machine learning model, and the training process is performed based on a plurality of joined training images. The joined training images are generated by performing the pre-processing process on a plurality of movie images with black bars and a plurality of non-movie images without black bars.
    Type: Application
    Filed: December 15, 2021
    Publication date: September 15, 2022
    Inventor: YEN-HSUN CHU
  • Publication number: 20220267923
    Abstract: A purification apparatus and a method of purifying hot zone parts are provided. The purification apparatus is configured to remove impurities attached on at least one hot zone part. The purification apparatus includes a crystal high temperature furnace, an enclosed box disposed in the crystal high temperature furnace, an outer tube connected to the crystal high temperature furnace and the enclosed box, an inner tube disposed in the outer tube, and a gas inlet cover connected to the outer tube. The crystal high temperature furnace includes a furnace body, a furnace cover, and a thermal field module disposed in the furnace body. The gas inlet cover is configured to input a noble gas into the enclosed box through the inner tube, and the thermal field module is configured to heat the noble gas so that the impurities are heated and vaporized through the noble gas.
    Type: Application
    Filed: February 24, 2022
    Publication date: August 25, 2022
    Inventors: Chung-Sheng Chang, Masami Nakanishi, YU-SHENG SU, YEN-HSUN CHU, YUNG-CHI WU, Yi-Hua Fan
  • Publication number: 20220238128
    Abstract: The audio processing method may include: utilizing a first convolution layer circuit and a second convolution layer circuit to perform convolution processing on an audio mixture signal, respectively, in order to generate an input feature map; performing separation processing corresponding to a predetermined neural network model on the input feature map to obtain a first output feature map and a second output feature map respectively corresponding to a first type and a second type; and utilizing a first deconvolution layer circuit and a second deconvolution layer circuit to perform deconvolution processing on different portions of the first output feature map, respectively, and utilizing the first deconvolution layer circuit and the second deconvolution layer circuit to perform deconvolution processing on different portions of the second output feature map, respectively, in order to generate corresponding processing results, for generating multiple audio output signals.
    Type: Application
    Filed: May 25, 2021
    Publication date: July 28, 2022
    Inventors: Tai-Shih Chi, Ya-Ching Lai, Yih-Liang Shen, Yen-Hsun Chu, Chia-Chun Cheng