Abstract: A voice recognition method and system. A voice command can be received from a speech recognition unit associated with a device by a language switching module configured in association with the device. The voice command is recognized and processed into particular content to identify via a language database, a language associated with the device. The language can then be changed based on a detected language so that the language can be altered without reference to an instruction manual. In one scenario, a user walks toward the machine or device associated with the user device and speaks the desired/known language. The device “listens” to the voice, detects the language, and changes the user interface accordingly.

Abstract: Systems and techniques are provided for training a natural language processing model with information retrieval model annotations. A natural language processing model may be trained, through machine learning, using training examples that include part-of-speech tagging and annotations added by an information retrieval model. The natural language processing model may generate part-of-speech, parse-tree, beginning, inside, and outside label, mention chunking, and named-entity recognition predictions with confidence scores for text in the training examples. The information retrieval model annotations and part-of-speech tagging in the training example may be used to determine the accuracy of the predictions, and the natural language processing model may be adjusted. After training, the natural language processing model may be used to make predictions for novel input, such as search queries and potential search results.

Abstract: A system for distinguishing and identifying speech segments originating from speech of one or more relevant speakers in a predefined detection area. The system includes an optical system which outputs optical patterns, each representing audio signals as detected by the optical system in the area within a specific time frame; and a computer processor which receives each of the outputted optical patterns and analyses each respective optical pattern to provide information that enables identification of speech segments thereby, by identifying blank spaces in the optical pattern, which define beginning or ending of each respective speech segment.

Abstract: Speech recognition models are dynamically re-configurable based on user information, background information such as background noise and transducer information such as transducer response characteristics to provide users with alternate input modes to keyboard text entry. The techniques of dynamic re-configurable speech recognition provide for deployment of speech recognition on small devices such as mobile phones and personal digital assistants as well environments such as office, home or vehicle while maintaining the accuracy of the speech recognition.

Abstract: A speaker indexing device extracts a plurality of features from a speech signal on a frame-by-frame basis, models a distribution of first feature sets by a mixture distribution containing as many probability distributions as there are speakers, selects for each probability distribution either first feature sets located within a predetermined distance from the center of the probability distribution or a predetermined number of first feature sets in sequence starting from the first feature set closest to the center of the probability distribution, selects a second feature for the frame corresponding to the selected first feature sets as first training data for the speaker corresponding to the probability distribution and, using the first training data, trains a speaker model to be used to append to each frame identification information for identifying the speaker speaking in the frame.

Abstract: According to one embodiment, an electronic device includes a display controller and circuitry. The display controller displays a first object indicative of a first speaker, a first object indicative of a second speaker different from the first speaker, a second object indicative of a first speech period identified as a speech of the first speaker, and a second object indicative of a second speech period identified as a speech of the second speaker. The circuitry integrates the first speech period and the second speech period into a speech period of a same speaker when a first operation of associating the first object indicative of the first speaker with the first object indicative of the second speaker is operated.

Abstract: A speech-based system is configured to use its audio-based user interface to present various types of device status information such as wireless signal strengths, communication parameters, battery levels, and so forth. In described embodiments, the system is configured to understand spoken user requests for device and system status. For example, the user may speak a request to obtain the current wireless signal strength of the speech-based system. The speech-based system may respond by determining the signal strength and by playing speech or other sound informing the user of the signal strength. Furthermore, the system may monitor operational parameters to detect conditions that may degrade the user experience, and may report such conditions using generated speech or other sounds.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining hotword suitability. In one aspect, a method includes receiving speech data that encodes a candidate hotword spoken by a user, evaluating the speech data or a transcription of the candidate hotword, using one or more predetermined criteria, generating a hotword suitability score for the candidate hotword based on evaluating the speech data or a transcription of the candidate hotword, using one or more predetermined criteria, and providing a representation of the hotword suitability score for display to the user.

Abstract: An apparatus for processing an audio signal and method thereof are disclosed.

Abstract: An information signal reconstructor is configured to reconstruct, using aliasing cancellation, an information signal from a lapped transform representation of the information signal including, for each of consecutive, overlapping regions of the information signal, a transform of a windowed version of the respective region, wherein the information signal reconstructor is configured to reconstruct the information signal at a sample rate which changes at a border between a preceding region and a succeeding region of the information signal.

Abstract: In general, techniques are described for editing of higher-order ambisonic audio data. A device comprising a memory and one or more processors may be configured to perform the techniques. The memory may be configured to store spherical harmonic (SH) basis functions. The one or more processors may be configured to manipulate the SH basis functions associated with higher order ambisonics coefficients to alter a direction of an audio object represented by the higher order ambisonics coefficients.

Abstract: Provided are, among other things, systems, methods and techniques for detecting whether a transient exists within an audio signal. According to one representative embodiment, a segment of a digital audio signal is divided into blocks, and a norm value is calculated for each of a number of the blocks, resulting in a set of norm values for such blocks, each such norm value representing a measure of signal strength within a corresponding block. A maximum norm value is then identified across such blocks, and a test criterion is applied to the norm values. If the test criterion is not satisfied, a first signal indicating that the segment does not include any transient is output, and if the test criterion is satisfied, a second signal indicating that the segment includes a transient is output. According to this embodiment, the test criterion involves a comparison of the maximum norm value to a different second maximum norm value, subject to a specified constraint, within the segment.

Abstract: Linear prediction based audio coding is improved by coding a spectrum composed of a plurality of spectral components using a probability distribution estimation determined for each of the plurality of spectral components from linear prediction coefficient information. The linear prediction coefficient information is available anyway. Accordingly, it may be used for determining the probability distribution estimation at both encoding and decoding side. The latter determination may be implemented in a computationally simple manner by using, for example, an appropriate parameterization for the probability distribution estimation at the plurality of spectral components. The coding efficiency as provided by the entropy coding is compatible with probability distribution estimations as achieved using context selection, but its derivation is less complex.

Abstract: Provided is a speech/audio encoding apparatus with which it is possible to code a significant frequency domain region with high precision, and to enable high audio quality. A speech/audio encoding apparatus codes a linear prediction coefficient. A significant frequency domain region detection unit identifies a frequency domain region which is aurally significant from the linear prediction coefficient. A frequency domain region repositioning unit repositions the significant frequency domain region which is identified by the significant frequency domain region detection unit. A bit allocation computation unit determines a coding bit allocation on the basis of the significant frequency domain region which is repositioned by the frequency domain region repositioning unit.

Abstract: Described herein are systems, methods and apparatus for decoding in-band on-channel signals and extracting audio and data signals. Memory requirements are reduced by selectively filtering a bit stream of data in the signal so that services of interest which are encoded therein are processed. A single pool of memory may be shared between physical layer and data link layer processing. Memory in this pool may be allocated dynamically between processing of data at the physical and data link layers. When the available memory is not sufficient to support the required services, the dynamic allocation allows for graceful degradation.

Abstract: An adaptive equalization system that adjusts the spectral shape of a speech signal based on an intelligibility measurement of the speech signal may improve the intelligibility of the output speech signal. Such an adaptive equalization system may include a speech intelligibility measurement module, a spectral shape adjustment module, and an adaptive equalization module. The speech intelligibility measurement module is configured to calculate a speech intelligibility measurement of a speech signal. The spectral shape adjustment module is configured to generate a weighted long-term speech curve based on a first predetermined long-term average speech curve, a second predetermined long-term average speech curve, and the speech intelligibility measurement. The adaptive equalization module is configured to adapt equalization coefficients for the speech signal based on the weighted long-term speech curve.

Abstract: A method for speech restoration by an electronic device is described. The method includes obtaining a noisy speech signal. The method also includes suppressing noise in the noisy speech signal to produce a noise-suppressed speech signal. The noise-suppressed speech signal has a bandwidth that includes at least three subbands. The method further includes iteratively restoring each of the at least three subbands. Each of the at least three subbands is restored based on all previously restored subbands of the at least three subbands.

Abstract: A method for reconstructing at least one target signal comprises determining a first set of feature vectors from the input signal, the first set of feature vectors forming a non-negative input matrix; determining a second set of feature vectors, the second set of feature vectors forming a non-negative noise matrix; decomposing the input matrix into a sum of a first matrix and a second matrix, the first matrix representing a product of a non-negative bases matrix and a non-negative weight matrix, and the second matrix representing a combination of the noise matrix and a noise weight vector; and reconstructing the at least one target signal based on the non-negative bases matrix and the non-negative weight matrix.

Abstract: A nonlinear acoustic echo signal suppression system and method using a Volterra filter is disclosed. The nonlinear acoustic echo signal suppression system includes an acoustic echo signal estimator configured to estimate a nonlinear acoustic echo signal by using a Volterra filter in a frequency filter, and a near-end talker speech signal generator configured to generate a near-end talker speech signal, in which the nonlinear acoustic echo signal is suppressed, by using a gain function based on a statistical model.

Abstract: Provided are systems and methods for generating clean speech from a speech signal representing a mixture of a noise and speech. The clean speech may be generated from synthetic speech parameters. The synthetic speech parameters are derived based on the speech signal components and a model of speech using auditory and speech production principles. The modeling may utilize a source-filter structure of the speech signal. One or more spectral analyzes on the speech signal are performed to generate spectral representations. The feature data is derived based on a spectral representation. The features corresponding to the target speech according to a model of speech are grouped and separated from the feature data. The synthetic speech parameters, including spectral envelope, pitch data and voice classification data are generated based on features corresponding to the target speech.

Abstract: A novel audio ducking method that is aware of the loudness levels of the audio content is provided. The method specifies a minimum loudness separation between audio tracks that are designated as masters and audio tracks that are designated as slaves. The method attenuates the volume of the slave tracks in order to provide at least the minimum loudness separation between the slave tracks and the master tracks. The amount of attenuation for a slave is determined based on the loudness levels of the slave and of a master.

Abstract: The present invention relates to an encoding device and method, and a decoding device and method, and a program which enable music signals to be played with higher sound quality by expanding a frequency band. A band pass filter divides an input signal into multiple subband signals, a feature amount calculating circuit calculates feature amount using at least any one of the divided multiple subband signals and the input signal, a high-frequency subband power estimating circuit calculates an estimated value of high-frequency subband power based on the calculated feature amount, and a high-frequency signal generating circuit generates a high-frequency signal component based on the multiple subband signals divided by the band pass filter and the estimated value of the high-frequency subband power calculated by the high-frequency subband power estimating circuit.

Abstract: There is provided an information processing device including an input unit to which words are input, an analysis unit configured to analyze meanings of the respective words, an image generation unit configured to generate single images corresponding to the respective words, and a display control unit configured to control the images generated by the image generation unit to be displayed on a display unit.

Abstract: A system and method of creating a customized multi-media message to a recipient is disclosed. The multi-media message is created by a sender and contains an animated entity that delivers an audible message. The sender chooses the animated entity from a plurality of animated entities. The system receives a text message from the sender and receives a sender audio message associated with the text message. The sender audio message is associated with the chosen animated entity to create the multi-media message. The multi-media message is delivered by the animated entity using as the voice the sender audio message wherein the mouth movements of the animated entity conform to the sender audio message.

Abstract: The invention relates to the analysis of characteristics of audio and/or video signals for the generation of audio-visual content signatures. To determine an audio signature a region of interest for example of high entropy—is identified in audio signature data. This region of interest is then provided as an audio signature with offset information. A video signature is also provided.

Abstract: Systems and techniques are provided for finding differences in nearly-identical audio recordings. A first version of an audio recording may be received. A second version of the audio recording may be received. A difference between the first version of the audio recording and the second version of the audio recording may be determined using time domain analysis and frequency domain analysis. The difference may be stored in a difference set. The difference set may allow the first version of the audio recording to be distinguished from the second version of the audio recording. The audio recording may be a music track. The first version of the audio recording may be an explicit version of the music track. The second version of the audio recording may be an edited version of the music track.

Abstract: A speaker change detection device sets first and second analysis periods before and after each of time points in a voice signal, generates, for each of the time points, a first speaker model from a distribution of features in frames in the first analysis period, and a second speaker model from a distribution of features in frames in the second analysis period, calculates, for each of the time points, a matching score representing the likelihood of similarity of features between a group of speakers in the first analysis period and a group of speakers in the second analysis period by applying the features extracted from the second analysis period to the first speaker model and applying the features extracted from the first analysis period to the second speaker model, and detects a speaker change point on the basis of the matching scores at the plurality of time points.

Abstract: Disclosed herein is a magnetic write head comprising a main pole, a write gap over the main pole, a first hot seed layer over the main pole, wherein the first hot seed layer comprises a first hot seed material having a first anisotropy value, and a second hot seed layer comprising a second hot seed material having a second anisotropy value, the second anisotropy value being greater than the first anisotropy value, wherein at least a portion of the second hot seed layer is adjacent to the first hot seed layer to mitigate adjacent track interference. Also disclosed are methods to fabricate magnetic write heads having first and second hot seed layers.

Abstract: A multi-sensor reader that includes a first sensor that has a sensor stack, which includes a free layer (FL) that has a magnetization that changes according to an external magnetic field. The first sensor also includes a shielding structure that is positioned over the sensor stack. The multi-sensor reader also includes a second sensor stacked over the first sensor. The second sensor includes a sensor stack, which includes a FL that has a magnetization that changes according to the external magnetic field. The multi-sensor reader further includes an isolation layer between the first sensor and the second sensor. A FL-to-FL spacing reduction feature is included in at least one of the isolation layer or the shielding structure.

Abstract: A cross-track profile of a known good test track on a recording medium is read during operation of a data storage device. Two or more center locations of the test track are determined at two or more different amplitude levels via a read transducer. A health condition of the read transducer is determined by comparing the two or more center locations of the test track with one another.

Abstract: A laser processing method processes spots on an object in a scan area with a laser beam emitted from a laser emitter. The method includes steps of covering the scan area in such a way as to transmit the laser beam and define a closed space over the scan area, jetting an inert gas into the closed space so that the jetted inert gas forms a gas curtain sweeping along the scan area in the closed space, and discharging the inert gas from the closed space while maintaining a positive pressure in the closed space, thereby removing plumes from the spots processed with the laser beam and preventing the spots from oxidizing. The method is capable of stably laser-processing the object without changing the color of the processed spots.

Abstract: Examples of slide-in ramps for hard-disk drives are disclosed. In one example according to aspects of the present disclosure, a slide-in ramp includes: a ramp body, a head load/unload section connected to the ramp body, and a slot defined in the ramp body. The slot includes a first portion and a second portion. A first upper surface of the ramp body surrounds the first portion of the slot at a top side of the ramp body, and a second upper surface of the ramp body surrounds the second portion of the slot at the top side of the ramp body. The second upper surface is non-coplanar with the first upper surface according to various examples.

Abstract: First and second controllers output respective first and second control signals for first and second plants that respectively correspond to the stages of an actuator. An inverse model of the first plant used to determine a decoupled output of the second plant from a combined output of the first and second plants. A disturbance estimate is subtracted from the second control signal to obtain a disturbance-corrected output. A loop-shaping filter is used between the disturbance corrected output and the second plant.

Abstract: A magnetic head includes a read transducer and a write transducer at a media-facing surface of the magnetic head. The magnetic head includes at least one heater that causes heat deformation at the media-facing surface in response to different first and second energizing currents. The first energizing current results in a first close point between the media-facing surface and a recording medium. The second energizing current results in a second close point between the media-facing surface and the recording medium. The second close point is at a different location in the media-facing surface than the first close point.

Abstract: An apparatus comprises a slider and an optical waveguide formed in the slider and configured to receive light from a laser source. A near-field transducer (NFT) is formed on the slider at or near an air bearing surface (ABS) of the slider and optically coupled to the waveguide. A bolometric sensor is positioned proximate the NFT and exposed to at least some of the light. The bolometric sensor is configured to detect changes in output optical power of the laser source and contact between the slider and a magnetic recording medium.

Abstract: A heat assisted magnetic recording (HAMR) assembly includes a slider, a laser diode and solder connections between the laser diode and the slider. The solder connections mechanically and electrically attach the laser diode to the slider. Each solder connection has a total volume per unit area (i.e., height) of less than or equal to about 15 ?m. The solder connections have a first intermetallic zone adjacent to the laser diode, a second intermetallic zone adjacent to the slider, and a eutectic zone of eutectic material between the first and second intermetallic zones. The eutectic zone occupies greater than or equal to about 35% of the total volume per unit area of the solder connection.

Abstract: A method for content playback and recording may include using a computer to obtain media content from a recorded medium. Concurrently with obtaining the media content, the method may include reencrypting the encrypted media content using a secondary encryption key and storing the reencrypted media content in a storage device.

Abstract: Techniques for reducing the time required for erasing specific data recorded on a tape medium. A specific group of records is erased without preliminarily locating the erasure end position. This is carried out by simultaneously utilizing three heads, that is, two read heads and one write head, to detect the erasure end position during data erasure. Various embodiments are applicable to tape media as well as other storage media. Various embodiments are not only applicable as a file system cooperating as a combination of hardware (H/W) and software (S/W), but also applicable in systems, such as databases, that directly use storage without an intermediary file system.

Abstract: A laser power applied to a recording head is changed for a plurality of iterations. Each iteration involves, via the recording head at each laser power, writing multiple adjacent tracks to a heat-assisted recording medium and determining a bit error rate for at least one of the adjacent tracks at each laser power. A first laser power is found that achieves a minimum bit error rate of the iterations. An operational value of laser power that is smaller than the first laser power is used during operational recording to reduce adjacent track interference.

Abstract: Systems, devices, apparatuses, components, methods, and techniques for cadence determination and media content selection are provided. An example media-playback device comprises a media-output device that plays media content items, a cadence-acquiring device, and a cadence-based media content selection engine. The cadence-acquiring device includes an accelerometer and a cadence-determination engine configured to determine a cadence based on acceleration data captured by the accelerometer. The cadence-based media content selection engine is configured to identify a media content item based on the cadence determined by the cadence-determining engine and cause the media-output device to playback the identified media content item.

Abstract: A request for writing a first index information to a first partition of a serial storage medium of a serial system is received. The serial storage medium includes a first partition for writing the first index information including a first index position, a second partition for storing a data set, the data set including a second index information including a second index position, a memory for determining the second index position, and a module for writing the first index position. A beginning position of the first index is written on the first partition. A determination is made whether the beginning of the first index position is proximate to the second index position. In response to the beginning of the first index position being proximate to the second index position, an interrupted index position is registered. A dummy index is written to the first partition at the interrupted index position.

Abstract: A read channel module including an input, a location module and a generation module. The input is configured to receive a read-back signal from a rotating storage medium. The location module is configured to determine a location to insert a first imitation defect within the read-back signal. The first imitation defect imitates a first defect. The generation module is configured to (i) selectively generate the first imitation defect, and (ii) insert the first imitation defect in the read-back signal at the determined location.

Abstract: Systems and methods are disclosed for detecting shingled overwrite errors. When a read error is encountered when reading from shingled recording tracks, a processor may determine whether the read error is an error caused by shingled overwriting. The processor may determine whether the read error is caused by shingled overwriting by determining read signal quality of one or more sectors preceding the read error, such as based on a bit error count or bit error ratio (BER), and comparing the read signal quality to a threshold value. The processor may determine that the read error is caused by shingled overwriting when the read signal quality value is lower than the threshold.

Abstract: Some embodiments of the invention provides a media editing application that includes tools to perform a variety of different editing operations based on roles assigned to media content. In some embodiments, the media editing application includes focus-editing tools to emphasize or de-emphasize different sets of clips based on the assigned roles. In some embodiments, the media editing application allows one or more sets of clips to be disabled or enabled during playback based on the assigned roles.

Abstract: A system and method for a media processor separates the functions of topology creation and maintenance from the functions of processing data through a topology. The system includes a control layer including a topology generating element to generate a topology describing a set of input multimedia streams, one or more sources for the input multimedia streams, a sequence of operations to perform on the multimedia data, and a set of output multimedia streams, and a media processor to govern the passing of the multimedia data as described in the topology and govern the performance of the sequence of multimedia operations on the multimedia data to create the set of output multimedia streams. The core layer includes the input media streams, the sources for the input multimedia streams, one or more transforms to operate on the multimedia data, stream sinks, and media sinks to provide the set of output multimedia streams.

Abstract: An imaging device includes a video acquisition unit, a tilt information acquisition unit, and an extraction period determination unit. The video acquisition unit obtains videos. The tilt information acquisition unit obtains information about shifts in posture, which are changes in angles of view or the like, during obtaining of the videos by the video acquisition unit. The extraction period determination unit determines periods to extract in the respective videos in accordance with the information obtained by the tilt information acquisition unit.

Abstract: In an aspect, in general, method for aligning an audio recording and a transcript includes receiving a transcript including a plurality of terms, each term of the plurality of terms associated with a time location within a different version of the audio recording, forming a plurality of search terms from the terms of the transcript, determining possible time locations of the search terms in the audio recording, determining a correspondence between time locations within the different version of the audio recording associated with the search terms and the possible time locations of the search terms in the audio recording, and aligning the audio recording and the transcript including updating the time location associated with terms of the transcript based on the determined correspondence.

Abstract: A method of operation of a display system includes: receiving a recap request for a source media; generating source content descriptors for the source media with a control unit; extracting a source recap segment from the source media based on correlation of the source content descriptors with a recap theme associated with the recap request; and generating a custom recap with the source recap segment for displaying on a device.

Abstract: Systems and methods for selectively recording and bookmarking a portion of broadcast media content include receiving a video stream containing the broadcast media content, determining a video segment of the video stream to be output to a display device, and performing OCR on characters present within the video segment. The systems and methods may further include detecting a trigger event in the OCR of the video segment and creating a digital bookmark corresponding to the detected trigger event. The systems and methods may include generating a recording of a portion of the broadcast media content, whereby the recording begins prior to the trigger event and concludes after the trigger event, and storing the digital bookmark associated with the generated recording.

Abstract: There are disclosed a method and a system for systematization of production-process video data. The method is executable at an electronic device.