Abstract: A temperature sensing device includes a substrate, a first reflective module, a first window cover, and a dual thermopile sensor. The first reflective module is disposed on the substrate, including a first mirror chamber with a narrow field of view (FOV), and the first reflective module focuses a thermal radiation from measured object to a first image plane in the first mirror chamber. The first window cover is disposed on the first reflective module, and the first window cover allows a selected band of the thermal radiation to pass through. The dual thermopile sensor is disposed on the substrate and located in the first mirror chamber, and the dual thermopile sensor senses a temperature data from the first image plane. Additional second reflective module, LED source plus pin hole with same FOV of dual thermopile sensor can illuminate the measured object for ease of placement of object to be heated.

Abstract: A semiconductor structure includes a dielectric layer over a substrate, a via conductor over the substrate and in the dielectric layer, and a first graphene layer disposed over the via conductor. In some embodiments, a top surface of the via conductor and a top surface of the dielectric layer are level. In some embodiments, the first graphene layer overlaps the via conductor from a top view. In some embodiments, the semiconductor structure further includes a second graphene layer under the via conductor and a third graphene layer between the dielectric layer and the via conductor. In some embodiments, the second graphene layer is between the substrate and the via conductor.

Abstract: A method for manufacturing a semiconductor structure includes forming a first interconnect feature in a first dielectric feature, the first interconnect feature including a first conductive element exposed from the first dielectric feature; forming a first cap feature over the first conductive element, the first cap feature including a first cap element which includes a two-dimensional material; forming a second dielectric feature with a first opening that exposes the first cap element; forming a barrier layer over the second dielectric feature while exposing the first cap element from the barrier layer; removing a portion of the first cap element exposed from the barrier layer; and forming a second conductive element in the first opening.

Abstract: A voice capturing method includes following operations: storing, by a buffer, voice data from a plurality of microphones; determining, by a processor, whether a target speaker exists and whether a direction of the target speaker changes according to the voice data and target speaker information; inserting a voice segment corresponding to a previous tracking direction into a current position in the voice data to generate fusion voice data when the target speaker exists and the direction of the target speaker changes from the previous tracking direction to a current tracking direction; performing, by the processor, a voice enhancement process on the fusion voice data according to the current tracking direction to generate enhanced voice data; performing, by the processor, a voice shortening process on the enhanced voice data to generate voice output data; and playing, by a playing circuit, the voice output data.

Abstract: A voice capturing method includes following operations: storing, by a buffer, voice data from a plurality of microphones; determining, by a processor, whether a target speaker exists and whether a direction of the target speaker changes according to the voice data and target speaker information; inserting a voice segment corresponding to a previous tracking direction into a current position in the voice data to generate fusion voice data when the target speaker exists and the direction of the target speaker changes from the previous tracking direction to a current tracking direction; performing, by the processor, a voice enhancement process on the fusion voice data according to the current tracking direction to generate enhanced voice data; performing, by the processor, a voice shortening process on the enhanced voice data to generate voice output data; and playing, by a playing circuit, the voice output data.

Abstract: A device for sound localization includes a spatial feature generator, a voice detector, an angle selector, and an angle retriever. The spatial feature generator generates M spatial feature signals according to signals of N microphones of a microphone array. The voice detector generates at least one voice detection signal according to at least one of the signals of the N microphones. The angle selector outputs a candidate angle signal according to the M spatial feature signals to indicate a candidate direction of sound. The angle retriever generates a sound detection result according to the M spatial feature signals to indicate whether any sound source exists, and then outputs an estimated angle signal indicative of a direction of sound according to the sound detection result, the at least one voice detection signal, and the candidate angle signal.

Abstract: Disclosed is a device for sound localization. The device can determine a direction of sound adequately, and includes a spatial feature generator, a voice detector, an angle selector, and an angle retriever. The spatial feature generator generates M spatial feature signals according to signals of N microphones of a microphone array. The voice detector generates at least one voice detection signal according to at least one of the signals of the N microphones. The angle selector outputs a candidate angle signal according to the M spatial feature signals to indicate a candidate direction of sound. The angle retriever generates a sound detection result according to the M spatial feature signals to indicate whether any sound source exists, and then outputs an estimated angle signal indicative of a direction of sound according to the sound detection result, the at least one voice detection signal, and the candidate angle signal.

Abstract: A method of audio source separation includes steps of applying a demixing matrix on a plurality of received signals to generate a plurality of separated results; performing a recognition operation on the plurality of separated results to generate a plurality of recognition scores; generating a constraint according to the plurality of recognition scores; and adjusting the demixing matrix according to the constraint; where the adjusted demixing matrix is applied to the plurality of received signals to generate a plurality of updated separated results from the plurality of received signals.

Abstract: The invention provides a wireless audiovisual information sharing system and method, applicable to a conference environment, wherein a user uses gesture touch to operate a user device at the at least one user end to perform device operation, and if the information receiving device of the information receiving end receives operation information of a user device of the at least one user end, the audiovisual signal outputted by the user device is decompressed and outputted until the operation information of another user device of the at least one user end is received and then the information receiving device decompresses and outputs the audiovisual signal outputted by the another user device, so that the users can share the wireless audiovisual information with each other through touch control.

Abstract: The invention provides a wireless audiovisual transmission system and method, applicable to conference presentation and/or real-time game environment. In each transmitting device of at least one transmitting end, a user can use gesture or voice to control each transmitting device, for operating the transmitting device, performing VIP setting or VIP deactivation. If the receiving device of the receiving end receives a VIP setting of a specific transmitting end of the at least one transmitting end, the receiving device preferentially decompresses the audiovisual signal outputted by the transmitting device of the specific transmitting end, and preferentially transmits decompressed the audiovisual signal to a display device for display, until the receiving device at the receiving end receives a VIP deactivation; if no VIP setting is received, the receiving device decompresses the at least a received audiovisual signal from at least one transmitting end according to a priority order for displaying.

Abstract: A method of audio source separation includes steps of applying a demixing matrix on a plurality of received signals to generate a plurality of separated results; performing a recognition operation on the plurality of separated results to generate a plurality of recognition scores; generating a constraint according to the plurality of recognition scores; and adjusting the demixing matrix according to the constraint; where the adjusted demixing matrix is applied to the plurality of received signals to generate a plurality of updated separated results from the plurality of received signals.

Abstract: In a speech recognition device and a speech recognition method, a key phrase containing at least one key word is received. The speech recognition method comprises steps: receiving a sound source signal of a key word and generating a plurality of audio signals; transforming the audio signals into a plurality of frequency signals; receiving the frequency signals to obtain a space-frequency spectrum and an angular estimation value thereof; receiving the space-frequency spectrum to define and output at least one spatial eigenparameter, and using the angular estimation value and the frequency signals to perform spotting and evaluation and outputting a Bhattacharyya distance; and receiving the spatial eigenparameter and the Bhattacharyya distance and using corresponding thresholds to determine correctness of the key phrase. Thereby this invention robustly achieves high speech recognition rate under very low SNR conditions.

Abstract: The present invention provides a spatially pre-processed target-to-jammer ratio weighted filter and a method thereof, which uses two microphones to receive audio signals. The audio signals are divided into a plurality of sinusoidal waves by a fast Fourier transform (FFT) module, and a beamformer uses the sinusoidal waves to generate beamformed signals. A reference generator generates at least one reference signal. The beamformed signals and reference signals are used to work out power spectral densities (PSD), and a target-to-jammer ratio (TJR) is worked out with the power spectral densities. TJR is used to determine whether a sound source exists. According to the determination result, a noise estimator is switched to eliminate noise from the beamformed signals and generate output signals. An inverse fast Fourier transform (IFFT) module recombines the output signals and then outputs the recombined signals.

Abstract: A speech recognition device and a speech recognition method thereof are disclosed. In the speech recognition method, a key phrase containing at least one key word is received. The speech recognition method comprises steps: receiving a sound source signal of a key word and generating a plurality of audio signals; transforming the audio signals into a plurality of frequency signals; receiving the frequency signals to obtain a space-frequency spectrum and an angular estimation value thereof; receiving the space-frequency spectrum to define and output at least one spatial eigenparameter and, and using the angular estimation value and the frequency signals to perform spotting and evaluation and outputting a Bhattacharyya distance; and receiving the spatial eigenparameter and the Bhattacharyya distance and using corresponding thresholds to determine correctness of the key phrase. Thereby this invention robustly achieves high speech recognition rate under very low SNR conditions.

Abstract: The present invention provides a spatially pre-processed target-to-jammer ratio weighted filter and a method thereof, which uses two microphones to receive audio signals. The audio signals are divided into a plurality of sinusoidal waves by a fast Fourier transform (FFT) module, and a beamformer uses the sinusoidal waves to generate beamformed signals. A reference generator generates at least one reference signal. The beamformed signals and reference signals are used to work out power spectral densities (PSD), and a target-to-jammer ratio (TJR) is worked out with the power spectral densities. TJR is used to determine whether a sound source exists. According to the determination result, a noise estimator is switched to eliminate noise from the beamformed signals and generate output signals. An inverse fast Fourier transform (IFFT) module recombines the output signals and then outputs the recombined signals.

Abstract: A power management method for a USB system comprises the steps of: detecting a connecting condition between the USB host and the USB device, reading a descriptor of the USB device, monitoring status of the USB host, selecting an operating mode of the USB device according to the status of the USB host, and selecting a corresponding transmission type via a USB port to transmit a control signal to the USB device so as to switch the operation mode of the USB device.

Abstract: A system for transmitting data in a virtual WAN comprises a plurality of routers, and each router comprises at least one connection port, a network status receiver and a redirection-processing unit. The network status receiver is configured to obtain bandwidth utilization information of other routers through the connection port. The redirection-processing unit is configured to determine whether the router needs to conduct a redirection and to establish a feasible redirection path based on the bandwidth utilization information of other routers.

Abstract: The present invention discloses a method of bandwidth management, which is used in network transmission, including two main steps of setting a user-level bandwidth and setting an application-level bandwidth. At the first step of setting a user-level bandwidth, firstly, the method provides a IP address for the master user, and secondly, provides at least one IP address for the at least one slave user, and then assigns a primary bandwidth to the IP address for the master user and at least one secondary bandwidth to at least one slave user's IP address accordingly. At the second step of setting an application-level bandwidth, the method executes at least one Internet application, and assigns at least one application bandwidth to at least one Internet application accordingly. At least one Internet application is provided. The setting and assignment of the primary bandwidth and the secondary bandwidth mentioned above could be performed in Graphics Interface, which is friendly to users.

Abstract: A synchronous multiple serial-lamps sets includes a plurality of serial or serial and parallel lamp sets; an IC set, an IC set auxiliary part, a power supply connector; and a signal transmission line. The leads connect the lamp sets, IC set, IC set auxiliary part, power supply connector so as to be formed as an integral structure. The power supply connector is connected to a power supply for supplying power to IC set and IC set auxiliary part. In operation, multiple functions are generated, and the lamp sets are operated so that the light emitters are illuminated or extinguished. Any one IC set transmits a signal to each IC set, so as to start or reset each lamp sets and thus they have the same preset electric functions, and thus a function of synchronous operation in a plurality of lamp sets is achieved.