Abstract: Presented herein are embodiments of a non-autoregressive sequence-to-sequence model that converts text to an audio representation. Embodiment are fully convolutional, and a tested embodiment obtained about 46.7 times speed-up over a prior model at synthesis while maintaining comparable speech quality using a WaveNet vocoder. Interestingly, a tested embodiment also has fewer attention errors than the autoregressive model on challenging test sentences. In one or more embodiments, the first fully parallel neural text-to-speech system was built by applying the inverse autoregressive flow (IAF) as the parallel neural vocoder. System embodiments can synthesize speech from text through a single feed-forward pass. Also disclosed herein are embodiments of a novel approach to train the IAF from scratch as a generative model for raw waveform, which avoids the need for distillation from a separately trained WaveNet.

Abstract: There is provided a method and system for carrying out two factor authentication, which renders an augmented reality environment or a virtual reality environment at the user device to depict an authentication object, and when a user interaction with the authentication object is detected, an authentication code received from an issuer server is displayed at the user device.

Abstract: Non-conductive substrates, especially the sidewalls of micro/nano holes thereof are chemically modified (i.e., chemically grafted) by reduced graphene oxide (rGO). The rGO possesses excellent electrical conductivity and therefore the modified substrates become conductive, so that it can be directly electroplated. These rGO-grafted holes can pass thermal shock reliability test after electroplating. The rGO grafting process possesses many advantages, such as a short process time, no complex agent (i.e., no chelator), no toxic agents (i.e., formaldehyde for electroless Cu deposition). It is employed in an aqueous solution instead of an organic solvent, and therefore is environmentally friendly and beneficial for industrial production.

Abstract: A method for manufacturing a copper foil with a rough surface in a plating tank includes causing an electrolyte solution to flow between an anode and a cathode with a current density of 5 ASF-40 ASF. The copper foil with a rough surface including dense nodules of single copper crystals is deposited on the cathode. The electrolyte solution includes chloride ions (20 ppm-80 ppm), polyethylene glycol (PEG) with a molecular weight of 400-8000 (100 ppm-700 ppm), sulfuric acid (20 g/L-200 g/L), copper sulfate pentahydrate (70 g/L-320 g/L) and a sulfur compound (1 ppm-60 ppm).

Abstract: The disclosure discloses a pixel structure. The pixel structure at least includes a first pixel electrode and a second pixel electrode disposed adjacently. The first pixel electrode includes a first effective display section and a first ineffective display section. The second pixel electrode includes a second effective display section and a second ineffective display section. A display area of the first effective display section is smaller than a display area of the second effective display section. A ratio of an actual coverage area of the second pixel electrode to an actual coverage area of the first pixel electrode is in a range of 1˜2/3.

Abstract: The method for producing a transparent conductive substrate includes forming metal meshes on a flexible non-conductive substrate with high transmittance. It's unnecessary to use palladium as a catalyst in this method. The metal meshes are in the form of nano/micro wires and the conductive substrate has high transmittance of 80%-90% at visible light wavelengths of 390-750 nm.

Abstract: A camera module comprises an image sensor and a lens module disposed on the image sensor. The lens module comprises a top glass structure at top of the lens module. The top glass structure includes a first glass substrate, a second glass substrate, and a baffle disposed immediately between the first and the second glass substrates. The top glass structure is an outermost layer of the camera module. The lens module also comprises a bottom glass substrate at bottom of the lens module. The bottom glass substrate is disposed on the image sensor.

Abstract: An optical source comprises a first laser arranged to generate a first optical signal having a first state of polarization (SOP) and a first optical frequency; a second laser arranged to generate a second optical signal having a second SOP, substantially orthogonal to the first SOP, and having a second optical frequency, different from the first optical frequency by a preselected frequency difference; a polarisation beam coupler arranged to combine the first optical signal and the second optical signal into a composite optical signal comprising both the first optical signal and the second optical signal having said substantially orthogonal SOPs; and an output arranged to output the composite optical signal.

Abstract: A wafer-level method for packaging a plurality of camera modules includes (a) overmolding a first housing material around a plurality of image sensors to produce a first wafer of packaged image sensors, (b) seating a plurality of lens units in the first wafer above the plurality of image sensors, respectively, and (d) overmolding a second housing material over the first wafer and around the lens units to form a second wafer of packaged camera modules, wherein each of the packaged camera modules includes one of the image sensors and one of the lens units, and the second housing material cooperates with the first housing material to secure the lens units in the second wafer.

Abstract: A method for packaging applies to packaging a plurality of wafer-level lenses. Each wafer-level lens includes (a) a substrate with opposite facing first and second surfaces and (b) a respective lens element on at least one of the first and second surfaces. Each lens element has a lens surface facing away from the substrate. The method includes partially encasing the plurality of wafer-level lenses with a housing material to produce a wafer of packaged wafer-level lenses. In the wafer of packaged wafer-level lenses, the housing material supports each of the plurality of wafer-level lenses by contacting the respective substrate, and the housing is shaped to form a plurality of housings for the plurality of wafer-level lenses, respectively.

Abstract: An electronic package is provided, which includes: a carrier; a plurality of electronic elements disposed on the carrier; a bather frame disposed on the carrier and positioned between adjacent two of the electronic elements; an encapsulant formed on the carrier and encapsulating the electronic elements and the bather frame with a portion of the bather frame protruding from the encapsulant; and a shielding element disposed on the encapsulant and being in contact with the portion of the bather frame protruding from the encapsulant. Therefore, the electronic package has an electromagnetic interference (EMI) shielding effect improved. The present disclosure further provides a method for fabricating the electronic package.

Abstract: Voice cloning is a highly desired capability for personalized speech interfaces. Neural network-based speech synthesis has been shown to generate high quality speech for a large number of speakers. Neural voice cloning systems that take a few audio samples as input are presented herein. Two approaches, speaker adaptation and speaker encoding, are disclosed. Speaker adaptation embodiments are based on fine-tuning a multi-speaker generative model with a few cloning samples. Speaker encoding embodiments are based on training a separate model to directly infer a new speaker embedding from cloning audios, which is used in or with a multi-speaker generative model. Both approaches achieve good performance in terms of naturalness of the speech and its similarity to original speaker—even with very few cloning audios.

Abstract: A method includes receiving a request for enrollment from a user at a wallet service provider (WSP) computer, and enrolling the user in the WSP computer. A message is transmitted to the user to query the user about accepting payment card account system transactions. A response to the message, from the user, is received in the WSP computer. The WSP computer downloads, to a device operated by the user, a list of acquirer financial institutions. The WSP computer receives, from the device operated by the user, an indication that the user has selected one of the acquirer financial institutions from the list. An application program supplied by the selected acquirer financial institution is downloaded to the device operated by the user.

Abstract: A method for facilitating a transaction with a one-time number includes: receiving a first transaction request, wherein the first transaction request includes a plurality of data elements including at least a first data element configured to store a zero transaction amount and a second data element configured to store a primary account number; parsing the primary account number stored in the second data element included in the received first transaction request; generating a one-time value, wherein the one-time value includes a predetermined number of digits and a subset of the predetermined number of digits is a reference to the processing server; storing a data entry comprised of at least the parsed primary account number and the generated one-time value; and transmitting the generated one-time value in response to the received first transaction request.

Abstract: In an optical transceiver, an optical transmitter coupled to a reconfigurable optical channel-add apparatus has first and second add paths, an add micro-ring resonator, and first and second optical attenuators, reconfigurable to selectively block an optical channel from an optical transmitter in one of the first and second add paths. The add micro-ring resonator is reconfigurable selectively to add an optical channel from the first add path to an optical waveguide to travel towards the first add-drop port or to add an optical channel from the second add path to the optical waveguide to travel towards the second add-drop port. An optical receiver is coupled to a reconfigurable optical channel-drop apparatus having a drop micro-ring resonator, and first and second drop paths.

Abstract: A structure light module comprises: a VCSEL substrate comprising a VCSEL array comprising a plurality of individual VCSELs; a first spacer disposed on the VCSEL substrate; a wafer level lens comprising a glass substrate and at least a replicated lens on a first surface of the glass substrate disposed on the first spacer; a second spacer disposes on the wafer level lens; a DOE disposed on the second spacer, where a structure light is projected from the DOE on a target surface for 3D imaging.

Abstract: Described herein are embodiments of an end-to-end text-to-speech (TTS) system with parallel wave generation. In one or more embodiments, a Gaussian inverse autoregressive flow is distilled from an autoregressive WaveNet by minimizing a novel regularized Kullback-Leibler (KL) divergence between their highly-peaked output distributions. Embodiments of the methodology computes the KL divergence in a closed-form, which simplifies the training process and provides very efficient distillation. Embodiments of a novel text-to-wave neural architecture for speech synthesis are also described, which are fully convolutional and enable fast end-to-end training from scratch. These embodiments significantly outperform the previous pipeline that connects a text-to-spectrogram model to a separately trained WaveNet. Also, a parallel waveform synthesizer embodiment conditioned on the hidden representation in an embodiment of this end-to-end model were successfully distilled.

Abstract: A data unit processing method performed by a first communication device, comprising: configuring a data unit by the first communication device, the data unit comprising: a specific format preamble following a first packet format; and a symbol part, comprising at least one symbol and following a Wi-Fi long range packet format. The specific format preamble can be decoded according to a first communication standard and the symbol can be decoded according to a Wi-Fi long range mode. The symbol part can be identified according to the specific format preamble. A method can decode the above-mentioned data unit and a communication device can perform the methods are also disclosed.

Abstract: The present embodiments provide a gradient coil assembly. The gradient coil assembly includes an aluminum wire body, and a copper wire end connected by cold pressure welding to two ends of the aluminum wire body. Using the gradient coil according to a particular embodiment, it is possible to reduce gradient coil weight as well as reduce the thickness of an outer vacuum chamber used for a magnet, thereby reducing the cost of the magnet and gradient coil, and making it less difficult to install and maintain the magnet and gradient coil. There is no problem of oxidation associated with the cold pressure welding of the aluminum wire body to the copper wire ends, so quality defects arising from such oxidation are avoided.

Abstract: A display pixel structure includes a plurality of pixel units arranged in an array. Each of the pixel units comprises the first sub-pixel, the second sub-pixel, the third sub-pixel and the fourth sub-pixel. Each of the sub-pixels includes a TFT switch. A ratio between a width/length ratio of the TFT switch of one of the first sub-pixel, the second sub-pixel, and the third sub-pixel or an average of width/length ratios of multiple TFT switches thereof and a width/length ratio of the TFT switch of the fourth sub-pixel is equal to a ratio between a storage capacitance of one of the first sub-pixel, the second sub-pixel, and the third sub-pixel or an average of multiple storage capacitances thereof and a storage capacitance of the fourth sub-pixel. An array substrate and a liquid crystal display device are also disclosed.