Convolutional neural networks (CNN) have shown promising results for end-to-end speech recognition, albeit still behind other state-of-the-art methods in performance. In this paper, we study how to bridge this gap and go beyond with a novel CNN-RNN-transducer architecture, which we call ContextNet. ContextNet features a fully convolutional encoder that incorporates global context information into convolution layers by adding squeeze-and-excitation modules. In addition, we propose a simple scaling method that scales the widths of ContextNet that achieves good trade-off between computation and accuracy. We demonstrate that on the widely used LibriSpeech benchmark, ContextNet achieves a word error rate (WER) of 2.1\%/4.6\% without external language model (LM), 1.9\%/4.1\% with LM and 2.9\%/7.0\% with only 10M parameters on the clean/noisy LibriSpeech test sets. This compares to the previous best published system of 2.0\%/4.6\% with LM and 3.9\%/11.3\% with 20M parameters. The superiority of the proposed ContextNet model is also verified on a much larger internal dataset.

Cognitive radio is a promising technology to improve spectral efficiency. However, the secure performance of a secondary network achieved by using physical layer security techniques is limited by its transmit power and channel fading. In order to tackle this issue, a cognitive unmanned aerial vehicle (UAV) communication network is studied by exploiting the high flexibility of a UAV and the possibility of establishing line-of-sight links. The average secrecy rate of the secondary network is maximized by robustly optimizing the UAV's trajectory and transmit power. Our problem formulation takes into account two practical inaccurate location estimation cases, namely, the worst case and the outage-constrained case. In order to solve those challenging non-convex problems, an iterative algorithm based on $mathcal{S}$-Procedure is proposed for the worst case while an iterative algorithm based on Bernstein-type inequalities is proposed for the outage-constrained case. The proposed algorithms can obtain effective suboptimal solutions of the corresponding problems. Our simulation results demonstrate that the algorithm under the outage-constrained case can achieve a higher average secrecy rate with a low computational complexity compared to that of the algorithm under the worst case. Moreover, the proposed schemes can improve the secure communication performance significantly compared to other benchmark schemes.

An apparatus for recognizing a representative user behavior includes a unit-data extracting unit configured to extract at least one unit data from sensor data, a feature-information extracting unit configured to extract feature information from each of the at least one unit data, a unit-behavior recognizing unit configured to recognize a respective unit behavior for each of the at least one unit data based on the feature information, and a representative-behavior recognizing unit configured to recognize at least one representative behavior based on the respective unit behavior recognized for each of the at least one unit data.

Apparatus and methods are provided for using automatic speech recognition to analyze a voice interaction and verify compliance of an agent reading a script to a client during the voice interaction. In one aspect of the invention, a communications system includes a user interface, a communications network, and a call center having an automatic speech recognition component. In other aspects of the invention, a script compliance method includes the steps of conducting a voice interaction between an agent and a client and evaluating the voice interaction with an automatic speech recognition component adapted to analyze the voice interaction and determine whether the agent has adequately followed the script. In yet still further aspects of the invention, the duration of a given interaction can be analyzed, either apart from or in combination with the script compliance analysis above, to seek to identify instances of agent non-compliance, of fraud, or of quality-analysis issues.

A device may include a physical phenomenon detector. The physical phenomenon detector may detect a physical phenomenon related to the device. In response to detecting the physical phenomenon, the device may record audio data that includes speech. The speech may be transcribed with a speech recognition engine. The speech recognition engine may be included in the device, or may be included with a remote computing device with which the device may communicate.

The present invention relates to a method for speaker recognition, comprising the steps of obtaining and storing speaker information for at least one target speaker; obtaining a plurality of speech samples from a plurality of telephone calls from at least one unknown speaker; classifying the speech samples according to the at least one unknown speaker thereby providing speaker-dependent classes of speech samples; extracting speaker information for the speech samples of each of the speaker-dependent classes of speech samples; combining the extracted speaker information for each of the speaker-dependent classes of speech samples; comparing the combined extracted speaker information for each of the speaker-dependent classes of speech samples with the stored speaker information for the at least one target speaker to obtain at least one comparison result; and determining whether one of the at least one unknown speakers is identical with the at least one target speaker based on the at least one comparison result.

A speech recognition system, method of recognizing speech and a computer program product therefor. A client device identified with a context for an associated user selectively streams audio to a provider computer, e.g., a cloud computer. Speech recognition receives streaming audio, maps utterances to specific textual candidates and determines a likelihood of a correct match for each mapped textual candidate. A context model selectively winnows candidate to resolve recognition ambiguity according to context whenever multiple textual candidates are recognized as potential matches for the same mapped utterance. Matches are used to update the context model, which may be used for multiple users in the same context.

A speech recognition method including the steps of receiving a speech input from a known speaker of a sequence of observations and determining the likelihood of a sequence of words arising from the sequence of observations using an acoustic model. The acoustic model has a plurality of model parameters describing probability distributions which relate a word or part thereof to an observation and has been trained using first training data and adapted using second training data to said speaker. The speech recognition method also determines the likelihood of a sequence of observations occurring in a given language using a language model and combines the likelihoods determined by the acoustic model and the language model and outputs a sequence of words identified from said speech input signal. The acoustic model is context based for the speaker, the context based information being contained in the model using a plurality of decision trees and the structure of the decision trees is based on second training data.

Various embodiments enable a device to perform tasks such as processing an image to recognize and locate text in the image, and providing the recognized text an application executing on the device for performing a function (e.g., calling a number, opening an internet browser, etc.) associated with the recognized text. In at least one embodiment, processing the image includes substantially simultaneously or concurrently processing the image with at least two recognition engines, such as at least two optical character recognition (OCR) engines, running in a multithreaded mode. In at least one embodiment, the recognition engines can be tuned so that their respective processing speeds are roughly the same. Utilizing multiple recognition engines enables processing latency to be close to that of using only one recognition engine.

When a speech sound of at least a predetermined sound pressure is externally input while a time measurement is not being performed, a time measuring circuit starts a time measurement responsive to a signal from a speech detector. When another speech sound of at least a predetermined sound pressure is externally input while a time measurement is being performed by the time measuring circuit, a measurement time measured by the time measuring circuit at this moment is stored in a time information memory. After a predetermined time has elapsed, if a speech recognition circuit recognizes that the externally input speech sound is a "stop" command, the time measurement operation performed by the time measuring circuit is stopped, and the time information stored in the time information memory is read out and displayed as measurement time information on a display unit.

A disposable electrode includes: an electrode pad; and a connector, connecting the electrode pad to a defibrillator, and including an information holder that can be provided with a transmissive opening or a light reflective member, the information holder holding information about at least an expiration date, depending on presence or absence of the transmissive opening or the light reflective member, the information holder allowing the information to be notified from the defibrillator when the connector is connected to the defibrillator.

The present invention relates to a method for selecting an appropriate mode when performing a new broadcast, such as a 3D stereo broadcast, a UHDTV broadcast, and a multi-view broadcast, among others, while maintaining compatibility with existing broadcasting channels in an MPEG-2-TS format for transmitting and receiving digital TV, and to a method for recognizing a descriptor. To this end, the present invention suggests providing the descriptor which is related to synthesizing left and right images using the type of stream, existence of the descriptor, and a frame-compatible mode flag.

There is provided a system and method for determining a number of players present using facial recognition. There is provided a method comprising capturing an image of the players present, and determining the number of players present based on the image. In this manner, players may more easily configure game settings, whereas spectators may be presented a more engaging experience.

An electronic device comprises a socket having a plurality of connection terminals that accommodates a card-type external device having a corner cut-out portion, a plurality of contact pads exposed on a surface of the card-type external device that are correspondingly connected to the connection terminals in response to insertion of the card-type external device into the socket. A detection unit detects erroneous insertion of the card-type external device into the socket in response to incorrect location of the cut-out portion during the erroneous insertion. A processor outputs a control signal in response to the detected erroneous insertion of the card-type external device.

A cognitive disorder diagnostic system that employs cognitive cubes, gameplay associate with the cognitive cubes, and a data gathering as statistical analysis base device that may be a computer, that communicates the gathered data to a web server host according to a unique ID associated with particular cognitive cubes and further associated with a particular player. Using the statistical data gathered using the gameplay, various cognitive disorders may be successfully diagnosed and treated with higher reliability.

A method and system are presented to address quantitative assessment of word recognition sensitivity of a subject, where the method comprises the steps of: (1) presenting at least one scene, comprising a plurality of letters and a background, to a subject on a display; (2) moving the plurality of letters relative to the scene; (3) receiving feedback from the subject via at least one; (4) quantitatively refining the received feedback; (5) modulating the saliency of the plurality of letters relative to accuracy of the quantitatively refined feedback; (6) calculating a critical threshold parameter; and (7) recording a critical threshold parameter onto a tangible computer readable medium.

The present invention provides method and apparatus to perform combined cognitive and motor rehabilitation on a computerized non-portable system or on single portable device. A patient can play a variety of games that require the patient to perform a variety of memory exercises which involve physical exertion. The activities of the patient are monitored with pattern analysis software which provides feedback to the patient. The feedback can include voice synthesis, video guidance, progression messages etc. Patient data obtained while the patient is performing each of the memory exercises is stored locally on a database module and then uploaded to a cloud server. A remote psychologist/psychiatrist monitors the patient by logging into the same cloud, and updating cognition exercises. The same therapist can have live chats with the patient for further interaction and coaching.

This invention provides a system and method for efficiently and completely combusting oil in mixture with particulate solids. A furnace (kiln) having a feed nozzle with a lead screw drives the mixture from a feed hopper. This nozzle includes forced-air jets/ports at its tip providing makeup air and allowing atomization of the mixture. The nozzle thereby directs the mixture into a rotating combustion chamber that is tilted downwardly from the front toward a solid waste outlet port at the rear. Uncombusted fuel and air backflow to an upper, secondary chamber near the primary chamber front, and are completely combusted at a high temperature. Gasses exit a flue that can include a heat exchanger. This heat exchanger can be operatively connected to a heating device or other mechanism that converts the heat into usable energy. The nozzle can include a cone with axially tilted air ports about its perimeter.

The amusement ride 1 comprises a track 2 and a vehicle 3 being moveable along the track 2 at a velocity v. Within the vehicle 3 a video camera 4 is installed. The video camera 4 takes a video film of the face of a passenger received within the vehicle 3 during a ride. A sender 5 transmits the data 6 to a facial expression recognition system 7. The result 10 of the process carried out by facial expression recognition system 7 may be downloaded from a server 11 by a client 13.

The bezel assembly for data reception, for use with a bill validator in a financial transactional device, includes a bezel housing and a data reception assembly. The bezel housing includes a customer-facing front portion and a back plate connectable to the bill validator that is mounted within the transactional device cabinet. The front portion includes an insertion/dispensing slot for receiving currency and a projecting protrusion forward of the casing. The forward-extending protrusion accommodates at least a portion of the data reception assembly. The bezel assembly can include a wireless communication function that is communicably connectable with a mobile device via a wireless communication method, a manual entry function, a biometric reader, one or more cameras for scanning and decrypting 2D barcodes and the like, thus enhancing the overall functionality of the financial transactional device.

An imaging part in a screen printing device which images a board and a screen mask includes a single camera which is disposed with a posture of horizontally facing towards an incidence optical axis, a half mirror which makes an imaging light, which is incident through a lower imaging optical axis, to be incident on a camera, and a mirror which makes an imaging light, which is incident through an upper imaging optical axis, to pass through the half mirror and to be incident on the camera, and further has an upper illuminating part and a lower illuminating part which individually illuminate respective imaging objects. Imaging light is taken in the camera in a state that the upper illuminating part and the lower illuminating part are individually operated in a mask imaging step and a board imaging step, respectively.

Methods, systems, and apparatus for receiving data indicating a location of a particular touchpoint representing a latest received touchpoint in a sequence of received touchpoints; identifying candidate characters associated with the particular touchpoint; generating, for each of the candidate characters, a confidence score; identifying different candidate sequences of characters each including for each received touchpoint, one candidate character associated with a location of the received touchpoint, and one of the candidate characters associated with the particular touchpoint; for each different candidate sequence of characters, determining a language model score and generating a transcription score based at least on the confidence score for one or more of the candidate characters in the candidate sequence of characters and the language model score for the candidate sequence of characters; selecting, and providing for output, a representative sequence of characters from among the candidate sequences of characters based at least on the transcription scores.

The present invention relates to the field of audio recognition, in particular to computer implemented note recognition methods in a gaming application. Furthermore, the present invention relates to improving latency of such audio recognition methods. One of the embodiments of the invention described herein is a method for note recognition of an audio source. The method includes: dividing an audio input into a plurality of frames, each frame having a pre-determined length, conducting a frequency analysis of at least a set of the plurality of frames, based on the frequency analysis, determining if a frame is a transient frame with a frequency change between the beginning and end of the frame, comparing the frequency analysis of each said transient frame to the frequency analysis of an immediately preceding frame and, based on said comparison, determining at least one probable pitch present at the end of each transient frame, and for each transient frame, outputting pitch data indicative of the probable pitch present at the end of the transient frame.

Established in 2000, Indigo Paints is a relatively new entrant to the decorative paints industry that is dominated by the like of Asian Paints, Berger and Nerolac.

Special CBI judge M P Chaudhary fixed June 21 for next hearing of the case.

CLAPPING the NHS each week is all well and good but surely we can think of a more permanent recognition?

I need help trying to find the font-family represented in this logo: PUNCH. Something very similar would be sufficient. Especially if it is a free font!Thanks for any suggestions!

This is a short one, and a change of approach. It’s a test run, really. (Every entry is an experiment of some sort.) Samples extracted from three different loops of the first-generation Buddha Machine, which dates from 2005, were recorded on the Teenage Engineering PO-33 K.O! and then run as a series of patterns, the […]

Join KUT’s Rebecca McInroy along with professors Art Markman and Bob Duke as they talk about the psychology of social activism, the effectiveness of deterrence, and the health consequences of negative emotions. Views and Brews is free and open to the public, hope to see you at the Cactus soon!

Dr. Art Markman and Dr. Bob Duke talk about some problems with research on embodied cognition and look at what it means and what it doesn’t.

Bradley and Karen interview Nick Coghlan, who works onn development and test infrastructure for Red Hat and is heavily involved with the Python community.

Show Notes:

Segment 0 (00:00:35)

Bradley and Karen interviewed Nick Coghlan who works for Red Hat and contributes to various Open Source and Free Software projects such as Python. Nick discussed his work on the infrastructure team at Red Hat, and his advocacy of Kallithea for use for the CPython project.

Send feedback and comments on the cast to <oggcast@faif.us>. You can keep in touch with Free as in Freedom on our IRC channel, #faif on irc.freenode.net, and by following Conservancy on on Twitter and and FaiF on Twitter.

Free as in Freedom is produced by Dan Lynch of danlynch.org. Theme music written and performed by Mike Tarantino with Charlie Paxson on drums.

The content of this audcast, and the accompanying show notes and music are licensed under the Creative Commons Attribution-Share-Alike 4.0 license (CC BY-SA 4.0).

IBM Cognos for Microsoft Office 11.0 Microsoft Windows 64bit Multilingual

IBM Cognos Analytics Server 11.1.5 Microsoft Windows Multilingual

IBM Cognos Analytics for Jupyter Notebook 11.1.6 Microsoft Windows Multilingual

IBM Cognos Transformer 11.0.0.68 Microsoft Windows Multilingual

IBM Cognos Analytics Client 11.1.5 Multiplatform Multilingual

IBM Cognos Analytics Installer 2.0.191205 Microsoft Windows Multilingual

IBM Cognos Analytics Client 11.1.6 Multiplatform Multilingual

IBM Cognos Analytics Server 11.1.6 Microsoft Windows Multilingual

IBM Cognos Analytics Installer 2.0.2003191 Microsoft Windows Multilingual

13-Apr-2015 17:37.

This item belongs to: audio/radio24.

This item has files of the following types: Archive BitTorrent, Metadata, VBR MP3

Constanza Henao, colombiana atrapada en la isla de San Martín

Nathan Cogswell, a junior out of Kentwood High, opened with a 6-under 65 in the first round Sunday on the 6,577-yard Pacific Dunes course. He slipped to a 72 in the second round Monday for a 5-under 137 total.

Nathan Cogswell, a junior out of Kentwood High, opened with a 6-under 65 in the first round Sunday on the 6,577-yard Pacific Dunes course. He slipped to a 72 in the second round Monday for a 5-under 137 total.

Learn how to integrate IT service management with AI services on an IoT device. You'll build a cognitive alert system for your IT operations.

Incognito’s joyous fourth album, full of smooth and authentic grooves.

When outback pioneering single mother Mary Emmott started rainfall records in 1914 she had no idea how important they would be.

A regional business lobby group says the Federal Government should recognise how important skilled foreign worker visas are in country areas.

Former Socceroos coach Ange Postecoglou claims a commanding win over Sydney FC in his long-awaited match-up against an A-League side as Yokohama F Marinos belt the Sky Blues in the Asian Champions League.

Humans get exercise in many ways, including training and competing with their dogs in agility trials. So is it time the competition is officially recognised as a sport?