Abstract: An apparatus for positioning a screen sound source, a method of generating loudspeaker set information for screen sound source positioning, and a method of reproducing a positioned screen sound source are provided. The apparatus and methods relate to a screen sound source positioning technique. A plurality of loudspeakers, each configured to have approximately the same gain, are each disposed proximate to the edge of a display, and a loudspeaker set including at least two of the loudspeakers is selected to position a virtual sound source substantially synchronized with a visual object displayed at a position on the screen of the display. Accordingly, a virtual sound source may be positioned at a certain specific position on the screen of a display without sound source distortion.

Abstract: A speech recognition apparatus is provided. The speech recognition apparatus includes a primary speech recognition unit configured to perform speech recognition on input speech and thus to generate word lattice information, a word string generation unit configured to generate one or more word strings based on the word lattice information, a language model score calculation unit configured to calculate bidirectional language model scores of the generated word strings selectively using forward and backward language models for each of words in each of the generated word strings, and a sentence output unit configured to output one or more of the generated word strings with high scores as results of the speech recognition of the input speech based on the calculated bidirectional language model scores.

Abstract: A dynamic response generating apparatus and method that may analyze an intention of a user based on user input information received from an inputting device, may analyze at least one of first response information with respect to the analyzed intention of the user, context information associated with the user input information, user motion information, and environmental information, may dynamically determine a modality with respect to the first response information, may process the first response information, and may dynamically generate second response information in a form of via the determined modality.

Abstract: An audio quality enhancing apparatus and method is provided in which a microphone array has a non-uniform configuration and thus a beam pattern of a desired direction is obtained in a wide range of frequencies including higher frequency bands and lower frequency bands even when the microphone array is relatively small.

Abstract: An apparatus and method for recognizing voice.

Abstract: An apparatus and method for recognizing a voice command for use in an interactive voice user interface are provided. The apparatus includes a command intention belief generation unit that is configured to recognize a first voice command and that may generate one or more command intention beliefs for the first voice command. The apparatus also includes a command intention belief update unit that is configured to update each of the command intention beliefs based on a system response to the first voice command and a second voice commands. The apparatus also includes a command intention belief selection unit that is configured to select one of the updated command intention beliefs for the first voice command. The apparatus also includes an operation signal output unit that is configured to select a final command intention from the selected updated command intention belief and to output an operation signal based on the selected final command intention.

Abstract: Disclosed is a golf ball including a center portion, at least two intermediate tiers sequentially surrounding the center portion, and an outer tier surrounding the intermediate tiers, which comprises: said outer tier having a thickness in a range of 0.8˜1.4 mm, and an a hardness (Shore D) in a range of 40˜50; among the intermediate tiers, said first intermediate tier formed just below the outer tier, having a thickness in a range of 1.0˜2.0 mm and a hardness (Shore D) in a range of 50˜65; each intermediate tier formed just below the first intermediate tier having a thickness in a range of 1.3˜1.5 mm, and a hardness (Shore D) in range of 30˜70; the radius of the center portion is in a range of 5˜18 mm, and a hardness (Shore D) in a range of 30˜70; and a moment of inertia value of the golf ball is less than 83 g·cm2. According to the golf ball of the present invention, an iron spin force is excellent as compared to a conventional golf ball for a match, and the flying distance is almost same.

Abstract: A flexible printed circuit board includes a first substrate portion having at least one first terminal, a second substrate portion in communication with the first substrate portion and having at least one circuit device, a connection substrate portion in communication with the second substrate portion, the connection substrate portion extending away from the second substrate portion in a same direction as the first substrate portion, and a third substrate portion in communication with the connection substrate portion, the third substrate portion having at least one second terminal.

Abstract: A user adaptive speech recognition method and apparatus is disclosed that controls user confirmation of a recognition candidate using a new threshold value adapted to a user. The user adaptive speech recognition method includes calculating a confidence score of a recognition candidate according to the result of speech recognition, setting a new threshold value adapted to the user based on a result of user confirmation of the recognition candidate and the confidence score of the recognition candidate, and outputting a corresponding recognition candidate as a result of the speech recognition if the calculated confidence score is higher than the new threshold value. Thus, the need for user confirmation of the result of speech recognition is reduced and the probability of speech recognition success is increased.

Abstract: A method and an apparatus for selecting a vocabulary closest to an input speech from among lexicons stored in memory, wherein a centroid lexicon representing lexicons belonging to a predetermined lexicon group is generated. Two lexicons, having a longest distance therebetween in the lexicon group, are selected using the centroid lexicon from the lexicon group, and a node indicating the lexicon group branches based on the two selected lexicons. A node having low group similarity is selected from among current terminal nodes, including branch nodes, and the above procedure is repeatedly performed on a lexicon group indicated by the selected node.

Abstract: Disclosed are an apparatus and method of deducing a user's intention using motion information. The user's intention deduction apparatus includes a speech intention determining unit configured to predict a speech intention regarding a user's speech using motion information sensed by at least one motion capture sensor, and a controller configured to control operation of detecting a voice section from a received sound signal based on the predicted speech intention.

Abstract: Disclosed are an apparatus and method of deducing a user's intention using multimodal information. The user's intention deduction apparatus includes a first predictor to predict a part of a user's intention using at least one piece of motion information, and a second predictor to predict the user's intention using the predicted part of the user's intention and multimodal information received from at least one multimodal sensor.

Abstract: A virtual screen sound source is spatially synchronized with a visual object displayed on a display. A plurality of loudspeaker sets, which each include at least three of a plurality of loudspeakers installed at the periphery of a display, are selected, individual sound sources corresponding to the respective selected loudspeaker sets are generated, and a multi-sound source is generated by overlapping the generated individual sound sources and output through loudspeakers included in the loudspeaker sets.

Abstract: An apparatus for positioning a screen sound source, a method of generating loudspeaker set information for screen sound source positioning, and a method of reproducing a positioned screen sound source are provided. The apparatus and methods relate to a screen sound source positioning technique. A plurality of loudspeakers, each configured to have approximately the same gain, are each disposed proximate to the edge of a display, and a loudspeaker set including at least two of the loudspeakers is selected to position a virtual sound source substantially synchronized with a visual object displayed at a position on the screen of the display. Accordingly, a virtual sound source may be positioned at a certain specific position on the screen of a display without sound source distortion.

Abstract: An apparatus and method for detecting a voice activity period. The apparatus for detecting a voice activity period includes a domain conversion module that converts an input signal into a frequency domain signal in the unit of a frame obtained by dividing the input signal at predetermined intervals, a subtracted-spectrum-generation module that generates a spectral subtraction signal which is obtained by subtracting a predetermined noise spectrum from the converted frequency domain signal, a modeling module that applies the spectral subtraction signal to a predetermined probability distribution model, and a speech-detection module that determines whether a speech signal is present in a current frame through a probability distribution calculated by the modeling module.

Abstract: An apparatus, medium, and method providing audio files with clustering, with audio files having information similar to queries input from a user being extracted and undergo clustering. A method for providing audio files with clustering includes calculating scores between queries input from a user and a specified audio file, detecting audio files having specified scores with the queries input from the user on the basis of the result of calculation and performing a dynamic clustering of the audio files, detecting the audio files having the specified scores with the queries input from the user and performing a static clustering of the audio files, and displaying the dynamic cluster or the static cluster on a screen.

Abstract: A method of automatic labeling using an optimum-partitioned classified neural network includes searching for neural networks having minimum errors with respect to a number of L phoneme combinations from a number of K neural network combinations generated at an initial stage or updated, updating weights during learning of the K neural networks by K phoneme combination groups searched with the same neural networks, and composing an optimum-partitioned classified neural network combination using the K neural networks of which a total error sum has converged; and tuning a phoneme boundary of a first label file by using the phoneme combination group classification result and the optimum-partitioned classified neural network combination, and generating a final label file reflecting the tuning result.

Abstract: Provided is a method and apparatus for searching music based on speech recognition. By calculating search scores with respect to a speech input using an acoustic model, calculating preferences in music using a user preference model, reflecting the preferences in the search scores, and extracting a music list according to the search scores in which the preferences are reflected, a personal expression of a search result using speech recognition can be achieved, and an error or imperfection of a speech recognition result can be compensated for.

Abstract: Provided are a multi-stage speech recognition apparatus and method. The multi-stage speech recognition apparatus includes a first speech recognition unit performing initial speech recognition on a feature vector, which is extracted from an input speech signal, and generating a plurality of candidate words; and a second speech recognition unit rescoring the candidate words, which are provided by the first speech recognition unit, using a temporal posterior feature vector extracted from the speech signal.

Abstract: Provided are a distributed speech recognition system, a distributed speech recognition speech method, and a terminal and a server for distributed speech recognition. The distributed speech recognition system includes a terminal which decodes a feature vector that is extracted from an input speech signal into a sequence of phonemes and generates the final recognition result by rescoring a candidate list provided from the outside; and a server which generates the candidate list by performing symbol matching on the recognized sequence of phonemes provided from the terminal and transmits the candidate list for the rescoring to the terminal.