Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for performing dynamic, stroke-based alignment of touch displays. In one aspect, a method includes obtaining a candidate transcription that an automated speech recognizer generates for an utterance, determining a particular context associated with the utterance, determining that a particular n-gram that is included in the candidate transcription is included among a set of undesirable n-grams that is associated with the context, adjusting a speech recognition confidence score associated with the transcription based on determining that the particular n-gram that is included in the candidate transcription is included among the set of undesirable n-grams that is associated with the context, and determining whether to provide the candidate transcription for output based at least on the adjusted speech recognition confidence score.

Abstract: Providing unique prompts for script, comprising sequencing, by a processor, at least one script in a design order having at least one prompt, identifying, by the processor, duplicate prompts of the at least one prompt, identifying, by the processor, unique prompts of the at least one prompt, translation tagging, by the processor, the identified unique prompts and replicating, by the processor, the translation tagged unique prompts to the identified duplicate prompts.

Abstract: Methods and systems for providing help prompts to a user of an automated dialog system are presented. In some embodiments, a computing device may receive a help request from the user of an automated dialog system. The help request may comprise a user request for information about one or more capabilities of the automated dialog system. The computing device may identify information expected to be input by the user to request that the automated dialog system perform its one or more capabilities. A natural language help prompt may be generated to provide guidance to the user to provide the identified information expected to be input.

Abstract: A method for analyzing a virtual discussion is provided. The method may include identifying, with a processing device, a first concept relevant to a first subdiscussion associated with an online discussion. The method may also include identifying a second concept relevant to the first subdiscussion, and determining a relation between the first concept and the second concept.

Abstract: A method for language prediction of a social network post includes generating a social network graph which includes nodes connected by edges. Some of the nodes are user nodes representing users of a social network and some of the nodes are social network post nodes representing social network posts. At least some of the users are authors of social network posts represented by respective social network post nodes. Edges of the graph are associated with respective weights. At least one of the social network post nodes is unlabeled. Language labels are predicted for the at least one unlabeled social network post node which includes propagating language labels through the graph. A language of the social network post is predicted based on the predicted language labels for the social network post node representing that social network post and optionally also based on content-based features.

Abstract: Methods and apparatus related to determining an edit rule based on a plurality of edits. Some implementations are directed to identifying the plurality of edits in one or more documents and determining an edit rule based on the pre-edit and post-edit phrases of the edits. Some implementations are directed to identifying the edits from one or more mature documents. The determined edit rule may be utilized to determine one or more candidate rephrasings of a subsequent phrase.

Abstract: A color information processing method includes: inputting reference data including color information about each pixel or each pixel group of a reference medium and evaluation target data including color information about each pixel or each pixel group of an evaluation medium by an input means; performing an image matching process of image matching the input reference data and evaluation target data; setting a specific area which is a unit of evaluation and includes a plurality of pixels to each data item subjected to the image matching process; and comparing the color information items about each pixel or each pixel group in the reference data and the evaluation target data in the set specific area to calculate a color difference and performing an averaging process to calculate an average evaluation result for the specific area by a calculation means.

Abstract: To detect events in an audio stream, frames of an audio signal (e.g., frames generated by a codec for a voice call or music stream) are received. Based on information in the frames, an index is used to look up an entry in a table associated with the codec. Each entry in the table indicates a likelihood that a frame matches a sound model element. The likelihood is used in the search for a sound bite, word, and/or phrase in the audio signal. The process of dynamic programming is used to find the combined likelihood for a match of the word, phrase, and/or sound bite to a region of the audio stream. Upon detection of the word, phrase, and/or sound bite in the audio stream, an event is generated, such as, notifying a person or logging the event in a database.

Abstract: A sound source-separating device includes a sound-collecting part, an imaging part, a sound signal-evaluating part, an image signal-evaluating part, a selection part that selects whether to estimate a sound source direction based on the first sound signal or the first image signal, a person position-estimating part that estimates a sound source direction using the first image signal, a sound source direction-estimating part that estimates a sound source direction, a sound source-separating part that extracts a second sound signal corresponding to the sound source direction from the first sound signal, an image-extracting part that extracts a second image signal of an area corresponding to the estimated sound source direction from the first image signal, and an image-combining part that changes a third image signal of an area other than the area for the second image signal and combines the third image signal with the second image signal.

Abstract: Mechanisms are provided for processing logical relationships in natural language content. A logical parse of a first parse of a natural language content is generated by identifying latent logical operators within the first parse indicative of logical relationships between elements of the natural language content. The logical parse comprises nodes and edges linking nodes. At least one knowledge value is associated with each node in the logical parse. The at least one knowledge value of at least a subset of the nodes in the logical parse is propagated to one or more other nodes in the logical parse based on propagation rules. A reasoning operation is performed on the logical parse to generate a knowledge output indicative of knowledge associated with one or more of the logical relationships between elements of the natural language content.

Abstract: A display apparatus which is capable of recognizing a voice and a method thereof are provided. The method includes receiving an uttered voice of a user, extracting a plurality of similar words which are similar to the uttered voice by extracting voice information from the uttered voice and measuring reliability of a plurality of words based on the extracted voice information, setting a word satisfying a predetermined condition from among the plurality of extracted similar words as a target word with respect to the uttered voice, and displaying at least one of the target word and a similar word list including similar words other than the target word. In this manner, a display apparatus may improve a recognition rate on an uttered voice of a user without changing an internal component related to voice recognition, such as, an acoustic model, a pronunciation dictionary, or the like.

Abstract: Disclosed herein is a noise cancellation apparatus and method, which select in advance parameters to be used for noise cancellation in a reference voice signal section by generating a reference voice signal in advance before a voice signal is generated, thus improving noise cancellation effects. The noise cancellation apparatus includes a parameter initialization unit for determining an initial value of a parameter to be used for noise cancellation, based on reference signals filtered for respective frequencies, a parameter estimation unit for receiving the initial value of the parameter, and estimating the parameter in response to signals that are input after being filtered for respective frequencies, a gain estimation unit for calculating gains for respective frequencies based on the parameter from the parameter estimation unit, and a gain application unit for cancelling noise by applying the gains to the signals that are input after being filtered for respective frequencies.

Abstract: A method of verifying the authenticity of an identification card, including capturing, with a point-of-sale scanner, a first image of the identification card, configuring an ultraviolet filter mounted within the point-of-sale scanner to allow the passage of ultraviolet light within, illuminating the identification card with ultraviolet light, capturing, with the point-of-sale scanner, a second image of the identification card, and determining if the identification card is authentic by analyzing the first and second images.

Abstract: A method for converting textual messages to musical messages comprising receiving a text input and receiving a musical input selection. The method includes analyzing the text input to determine text characteristics and analyzing a musical input corresponding to the musical input selection to determine musical characteristics. Based on the text characteristic and the musical characteristic, the method includes correlating the text input with the musical input to generate a synthesizer input, and sending the synthesizer input to a voice synthesizer. The method includes receiving a vocal rendering of the text input from the voice synthesizer, generating a musical message from the vocal rendering and the musical input, and outputting the musical message.

Abstract: Systems, methods, and computer program products to detect a keyword in speech, by generating, from a sequence of spectral feature vectors generated from the speech, a plurality of blocked feature vector sequences, and analyzing, by a neural network, each of the plurality of blocked feature vector sequences to detect the presence of the keyword in the speech.

Abstract: A value of gain is updated so that the greater the difference between the number of bits or estimated number of bits in a code obtained by encoding a string of integer value samples obtained by dividing each sample in a sample string derived from an input audio signal in a given interval by gain before the update and a predetermined number B of allocated bits, the greater the difference between the gain before the update and the updated gain. A gain code corresponding to the updated gain and an integer signal code obtained by encoding a string of integer value samples obtained by dividing each sample in the sample string by the gain are obtained.

Abstract: The present technique relates to a decoding device, a decoding method, an encoding device, an encoding method, and a program which can obtain a high-quality realistic sound. The encoding device stores speaker arrangement information in a comment region in a PCE of an encoded bit stream and stores a synchronous word and identification information in the comment region such that other public comments and the speaker arrangement information stored in the comment region can be distinguished from each other. When an encoded bit stream is decoded, it is determined whether the speaker arrangement information is stored on the basis of the synchronous word and the identification information stored in the comment region. Audio data included in the encoded bit stream is output according to the arrangement of the speakers corresponding to the determination result. The present technique can be applied to an encoding device.

Abstract: Methods and systems for managing and transmitting content are disclosed. A sample method can comprise determining signal-to-noise ratio information relating to one or more data blocks and determining a threshold signal-to-noise ratio. At least one of the one or more data blocks can be requested based upon respective signal-to-noise ratio information and the threshold signal-to-noise ratio.

Abstract: A systems, article, and method of automatic speech recognition using on-the-fly word lattice generation with word histories.

Abstract: Methods, systems, and computer-readable and executable instructions for spatial audio database based noise discrimination are described herein. For example, one or more embodiments include comparing a sound received from a plurality of microphones to a spatial audio database, discriminating a speech command and a background noise from the received sound based on the comparison to the spatial audio database, and determining an instruction based on the discriminated speech command.