Abstract: An electronic device is disclosed. The electronic device includes a substrate, a plurality of color filters disposed on the substrate, an optical film disposed on the plurality of color filter, and a defect disposed between the substrate and the optical film. The optical film has a first base, a protective layer on the first base, and a second base between the first base and the protective layer and having a first processed area. In a top view of the electronic device, the first processed area corresponds to the defect and at least partially overlaps at least two color filters.

Abstract: A channel hiatus correction method for an HDMI device is provided. A recovery code from scrambled data of the stream is obtained. A liner feedback shift register (LFSR) value of channels of the HDMI port is obtained based on the recovery code and the scrambled data of the stream. The stream is de-scrambled according to the LFSR value of the channels of the HDMI port. Video data is displayed according to the de-scrambled stream.

Abstract: An electrochemical equipment and system thereof for reduction of carbon dioxide is provided with cathode compartment, catholyte compartment, anode compartment, anolyte chamber, and isolation unit.

Abstract: An accelerator for neural network computing includes hardware engines and a buffer memory. The hardware engines include a convolution engine and at least a second engine. Each hardware engine includes circuitry to perform neural network operations. The buffer memory stores a first input tile and a second input tile of an input feature map. The second input tile overlaps with the first input tile in the buffer memory. The convolution engine is operative to retrieve the first input tile from the buffer memory, perform convolution operations on the first input tile to generate an intermediate tile of an intermediate feature map, and pass the intermediate tile to the second engine via the buffer memory.

Abstract: An accelerator for neural network computing includes hardware engines and a buffer memory. The hardware engines include a convolution engine and at least a second engine. Each hardware engine includes circuitry to perform neural network operations. The buffer memory stores a first input tile and a second input tile of an input feature map. The second input tile overlaps with the first input tile in the buffer memory. The convolution engine is operative to retrieve the first input tile from the buffer memory, perform convolution operations on the first input tile to generate an intermediate tile of an intermediate feature map, and pass the intermediate tile to the second engine via the buffer memory.

Abstract: The present invention discloses a complete speech recognition system having a training button and a recognition button, and the whole system uses the application specific integrated circuit (ASIC) architecture for the design, and also uses the modular design to divide the speech processing into 4 modules: system control module, autocorrelation and linear predictive coefficient module, cepstrum module, and DTW recognition module. Each module forms an intellectual product (IP) component by itself. Each IP component can work with various products and application requirements for the design reuse to greatly shorten the time to market.

Abstract: The present invention discloses a complete speech recognition system having a training button and a recognition button, and the whole system uses the application specific integrated circuit (ASIC) architecture for the design, and also uses the modular design to divide the speech processing into 4 modules: system control module, autocorrelation and linear predictive coefficient module, cepstrum module, and DTW recognition module. Each module forms an intellectual product (IP) component by itself. Each IP component can work with various products and application requirements for the design reuse to greatly shorten the time to market.