Abstract: Techniques for adapting a trained neural network acoustic model, comprising using at least one computer hardware processor to perform: generating initial speaker information values for a speaker; generating first speech content values from first speech data corresponding to a first utterance spoken by the speaker; processing the first speech content values and the initial speaker information values using the trained neural network acoustic model; recognizing, using automatic speech recognition, the first utterance based, at least in part on results of the processing; generating updated speaker information values using the first speech data and at least one of the initial speaker information values and/or information used to generate the initial speaker information values; and recognizing, based at least in part on the updated speaker information values, a second utterance spoken by the speaker.

Abstract: A stack comprises a substrate, a magnetic recording layer, and a negative thermal expansion layer disposed between the substrate and the magnetic recording layer. The negative thermal expansion layer is configured to reduce thermal profile changes of a surface of the stack opposite the substrate during a heat assisted magnetic recording write operation.

Abstract: The present invention relates to a quick quantitative analysis method and analyzer for mixture concentration information. A quick quantitative analysis method for a mixture based on spectral information includes: collecting a sample of a mixture solution, and injecting into a DAD to obtain DAD data; then, calculating a spectral library by a quick qualitative analysis method based on a vector error algorithm to obtain a spectral error vector; determining a spectral curve of a component included in the data corresponding to the sample, according to a value of the error vector; and calculating a content flag value of each component in the sample and concentration information of each substance.

Abstract: The disclosure provides a method, system, and apparatus for ticket checking based on a server-recommended ticket checking scheme. The method includes: receiving a ticket checking scheme recommendation request; responding to the ticket checking scheme recommendation request, retrieving a ticket checking scheme that is matched based on eligible data included in the ticket checking scheme recommendation request; matching the data included in the ticket checking scheme recommendation request with the ticket checking scheme; and setting and storing at least one form of password data for ticket checking in the ticket checking scheme after data matching. The disclosed embodiments reasonably recommend a ticket checking mode on the basis of existing venue characteristics, while ensuring the security of ticket checking. Ticket checking efficiency is thereby improved, and staff and equipment input and ticket checking costs are thereby reduced.

Abstract: The disclosed embodiments provide methods and systems for verifying a ticket at a client. One method comprises: sending a request for an electronic ticket verification interface to a server, the request including identification information of at least one purchased electronic ticket; receiving response information returned when the server determines by parsing and verification that electronic ticket information included in the request meets a ticket verification condition, and displaying, in a corresponding display box, an electronic ticket verification interface invoked by an electronic ticket verification interface invoking instruction included by the response information; and sending password input information collected on the verification interface to the server within a limit of password input times, and receiving ticket verification prompt information returned by the server in response to the password input information.

Abstract: Techniques for adapting a trained neural network acoustic model, comprising using at least one computer hardware processor to perform: generating initial speaker information values for a speaker; generating first speech content values from first speech data corresponding to a first utterance spoken by the speaker; processing the first speech content values and the initial speaker information values using the trained neural network acoustic model; recognizing, using automatic speech recognition, the first utterance based, at least in part on results of the processing; generating updated speaker information values using the first speech data and at least one of the initial speaker information values and/or information used to generate the initial speaker information values; and recognizing, based at least in part on the updated speaker information values, a second utterance spoken by the speaker.

Abstract: The disclosed embodiments provide a method, system, and device for online ticket checking based on a client interface. The method comprises: receiving a request for an electronic ticket checking interface sent by a client; sending response information to the client after parsing and verifying that identification information of at least one electronic ticket purchased by the client included in the request for an electronic ticket checking interface complies with a ticket checking condition; and receiving password input information acquired by a checking interface sent by the client, comparing whether the password input information is consistent with a password stored in a storage device A of a database, and sending, to the client, ticket checking prompt information based on the result of the comparison.

Abstract: Techniques for adapting a trained neural network acoustic model, comprising using at least one computer hardware processor to perform: generating initial speaker information values for a speaker; generating first speech content values from first speech data corresponding to a first utterance spoken by the speaker; processing the first speech content values and the initial speaker information values using the trained neural network acoustic model; recognizing, using automatic speech recognition, the first utterance based, at least in part on results of the processing; generating updated speaker information values using the first speech data and at least one of the initial speaker information values and/or information used to generate the initial speaker information values; and recognizing, based at least in part on the updated speaker information values, a second utterance spoken by the speaker.

Abstract: A system for providing a persona-driven and artificially-intelligent avatar is disclosed. The avatar may be utilized to represent an artificially-intelligent virtual assistant that may perform actions on behalf of a user, such as actions related to an investment strategy for the user. Notably, the avatar may influence artificial intelligence algorithms supporting the functionality of the virtual assistant via quantifiable traits of a persona of the avatar. In particular, the quantifiable traits of the persona of the avatar may be utilized to range bound the controlling parameters of the algorithms such that the virtual assistant represented by the avatar performs actions that conform to the ranged-bounded controlling parameters. Metrics corresponding to the performance of the virtual assistant may be tracked, and the system may modify parameters of the virtual assistant based on the metrics so as to adjust future actions performed by the virtual assistant for the user.

Abstract: Techniques for adapting a trained neural network acoustic model, comprising using at least one computer hardware processor to perform: generating initial speaker information values for a speaker; generating first speech content values from first speech data corresponding to a first utterance spoken by the speaker; processing the first speech content values and the initial speaker information values using the trained neural network acoustic model; recognizing, using automatic speech recognition, the first utterance based, at least in part on results of the processing; generating updated speaker information values using the first speech data and at least one of the initial speaker information values and/or information used to generate the initial speaker information values; and recognizing, based at least in part on the updated speaker information values, a second utterance spoken by the speaker.

Abstract: A method for transmitting information performed by a terminal is disclosed. The method includes: acquiring a first network identifier of a detected wireless communication network; sending a SIM card information request to a server, the SIM card information request including at least the first network identifier; receiving identity information and encryption data of a first SIM card sent by the server, the first SIM card corresponding to the detected wireless communication network corresponding to the first network identifier included in the SIM card information request; and joining the detected wireless communication network corresponding to the first network identifier using the identity information and encryption data of the first SIM card, and transmitting information through the detected wireless communication network.

Abstract: A method for transmitting information performed by a terminal is disclosed. The method includes: acquiring a first network identifier of a detected wireless communication network; sending a SIM card information request to a server, the SIM card information request including at least the first network identifier; receiving identity information and encryption data of a first SIM card sent by the server, the first SIM card corresponding to the detected wireless communication network corresponding to the first network identifier included in the SIM card information request; and joining the detected wireless communication network corresponding to the first network identifier using the identity information and encryption data of the first SIM card, and transmitting information through the detected wireless communication network.

Abstract: The invention relates to a perpendicular magnetic recording medium having a substrate and a seed layer comprising a Ni alloy.

Abstract: The invention relates to a perpendicular magnetic recording medium having a substrate and a seed layer comprising a Ni alloy.

Abstract: A scratch erasure resistant perpendicular magnetic recording medium comprises a non-magnetic substrate having a surface, and a layer stack formed over the surface and comprising: (i) at least one magnetically hard perpendicular magnetic recording layer; and (ii) at least one low shear modulus layer comprising at least one material having a shear modulus not greater than about 30 GPa. Preferably, the at least one magnetically hard perpendicular magnetic recording layer includes at least a first layer comprised of a magnetic material having a hexagonal close packed (hcp) crystal structure and <0001> preferred basal plane crystallographic orientation with c-axis perpendicular to a surface thereof.

Abstract: A magnetic recording medium is formed with a distribution of low coercivity regions functioning as a transition pattern for servo information capable of being sensed by a read/write head by exposing a masked magnetic layer to ions to change the coercivity of the exposed magnetic layer without substantially affecting the topography of the magnetic layer. Embodiments of the present invention include forming a series of substantially radially extending low coercivity regions used to divide the magnetic layer into a plurality of sectors comprising substantially concentric circumferentially extending data tracks by exposing a masked magnetic layer having a high coercivity, i.e. from about 2000 Oe to about 10000 Oe, to one or more heavy atom ion bombardments of gaseous ions, e.g. argon ions, at a dose of about 1×1013 atoms/cm2 to about 9×1015 atoms/cm2 having an implantation energy of about 10 KeV to about 50 KeV.

Abstract: A perfluoropolyether lubricant topcoat having at least one end-group having an epoxide ring is applied to a magnetic recording medium. Post-lubing treatment, such as ultraviolet (UV) light irradiation, of the perfluoropolyether lubricant topcoat initiates chemical bonding between the epoxide end-group and the medium surface. The lubricant of the present invention provides improved tribological performance over conventional lubricants.

Abstract: A method for measuring servo pattern irregularity and other servo information of pre-patterned servo media mounted on a spindle with the center of the pre-pattern tracks off center from the axis rotation of the spindle. An actuator is stepped radially while measuring the PES signals from the servo sectors of at least one highly eccentric track to form a cross track profile of the PES signals for each sector. A mathematical algorithm removes step measurement error from the data. A second mathematical algorithm removes non-repeatable random noise from the data. The amount of phase-in that radially aligns the respective cross track profiles is computed. Servo pattern written-in repeatable runout is determined by unwrapping the phase-in values to form a periodic curve.

Abstract: A servo-patterned magnetic recording medium, comprising: a magnetic layer having a surface with substantially uniform topography, the magnetic layer including a data zone and a servo pattern, the servo pattern comprising: (a) a first patterned plurality of regions of first, higher values of magnetic coercivity Hc and magnetic remanence-thickness product Mrt; and (b) a second patterned plurality of ion-implanted regions of second, lower values of Hc. and Mrt; wherein the second, lower values of Hc and Mrt are sufficiently lower than the first, higher values of Hc and Mrt as to permit sensing for enabling accurate positioning of a read/write transducer head in the data zone but sufficiently high for providing the medium with thermal stability, high amplitude of magnetic transition, and high signal-to-noise ratio.

Abstract: Disclosed is a method measuring the height of a magneto-resistive element (MRE) on a padded slider in a disc drive based on a change in the resistance of the MRE. During operation of the disc drive a biasing current is applied to the MRE head. Next a resistance value of the MRE head is calculated by determining the voltage drop across the MRE head. The resistance of the MRE head is dependent upon its temperature. The temperature of the MRE lowers as its proximity to the disc increases, as the cooler disc surface acts as a conduit drawing heat away from the MRE. This measured resistance value is compared to a threshold value. The threshold value is based upon the resistance of the MRE head at a threshold temperature, which corresponds to a specific distance away from the disc surface. If the measured resistance is lower than the threshold resistance the method outputs an indication to a user that the drive is in danger of failing.