Abstract: A computer-implemented method for generating an input for a classifier. The method includes obtaining n-best hypotheses which is an output of an automatic speech recognition (ASR) for an utterance, combining the n-best hypotheses horizontally in a predetermined order with a separator between each pair of hypotheses, and outputting the combined n-best hypotheses as a single text input to a classifier.

Abstract: Symbol sequences are estimated using a computer-implemented method including detecting one or more candidates of a target symbol sequence from a speech-to-text data, extracting a related portion of each candidate from the speech-to-text data, detecting repetition of at least a partial sequence of each candidate within the related portion of the corresponding candidate, labeling the detected repetition with a repetition indication, and estimating whether each candidate is the target symbol sequence, using the corresponding related portion including the repetition indication of each of the candidates.

Abstract: A dialog server which provides dialogs made by at least one user through their respective avatars in a virtual space. A method and a computer readable article of manufacture tangibly embodying computer readable instructions for executing the steps of the method are also provided. The dialog server includes: a position storage unit which stores positional information on the avatars; an utterance receiver which receives at least one utterance of avatars and utterance strength representing an importance or attention level of the utterance; an interest level calculator which calculates interest levels between avatars based on their positional information; a message processor which generates a message based on the utterance in accordance with a value calculated from the interest levels and the utterance strength; and a message transmitter which transmits the message to the avatars.

Abstract: Embodiments include methods and systems for improving an acoustic model. Aspects include acquiring a first standard deviation value by calculating standard deviation of a feature from first training data and acquiring a second standard deviation value by calculating standard deviation of a feature from second training data acquired in a different environment from an environment of the first training data. Aspects also include creating a feature adapted to an environment where the first training data is recorded, by multiplying the feature acquired from the second training data by a ratio obtained by dividing the first standard deviation value by the second standard deviation value. Aspects further include reconstructing an acoustic model constructed using training data acquired in the same environment as the environment of the first training data using the feature adapted to the environment where the first training data is recorded.

Abstract: Embodiments include methods and systems for improving an acoustic model. Aspects include acquiring a first standard deviation value by calculating standard deviation of a feature from first training data and acquiring a second standard deviation value by calculating standard deviation of a feature from second training data acquired in a different environment from an environment of the first training data. Aspects also include creating a feature adapted to an environment where the first training data is recorded, by multiplying the feature acquired from the second training data by a ratio obtained by dividing the first standard deviation value by the second standard deviation value. Aspects further include reconstructing an acoustic model constructed using training data acquired in the same environment as the environment of the first training data using the feature adapted to the environment where the first training data is recorded.

Abstract: A reading accuracy-improving system includes: a reading conversion unit for retrieving a plurality of candidate word strings from speech recognition results to determine the reading of each candidate word string; a reading score calculating unit for determining the speech recognition score for each of one or more candidate word strings with the same reading to determine a reading score; and a candidate word string selection unit for selecting a candidate to output from the plurality of candidate word strings on the basis of the reading score and speech recognition score corresponding to each candidate word string.

Abstract: Method of selecting training text for language model, and method of training language model using the training text, and computer and computer program for executing the methods. The present invention provides for selecting training text for a language model that includes: generating a template for selecting training text from a corpus in a first domain according to generation techniques of: (i) replacing one or more words in a word string selected from the corpus in the first domain with a special symbol representing any word or word string, and adopting the word string after replacement as a template for selecting the training text; and/or (ii) adopting the word string selected from the corpus in the first domain as the template for selecting the training text; and selecting text covered by the template as the training text from a corpus in a second domain different from the first domain.

Abstract: Method of selecting training text for language model, and method of training language model using the training text, and computer and computer program for executing the methods. The present invention provides for selecting training text for a language model that includes: generating a template for selecting training text from a corpus in a first domain according to generation techniques of: (i) replacing one or more words in a word string selected from the corpus in the first domain with a special symbol representing any word or word string, and adopting the word string after replacement as a template for selecting the training text; and/or (ii) adopting the word string selected from the corpus in the first domain as the template for selecting the training text; and selecting text covered by the template as the training text from a corpus in a second domain different from the first domain.

Abstract: A method and system are provided for training word embedding of domain-specific words. The method includes training, by a processor, a first word embedding, using a general domain corpus, on one or more terms inputted by a user. The method further includes retraining, by the processor, the first word embedding, using a specific domain corpus, for a Neuro-Linguistic Programming task, to create a tuned word embedding. The method also includes training, by the processor, a Neural Network for the Neuro-Linguistic Programming task, using the specific domain corpus. The method additionally includes incorporating, by the processor, the trained Neural Network and tuned word embedding into a Neural Network-based Neuro-Linguistic Programming task. The retraining of the first word embedding and the training of the Neural Network are performed together, and the tuned word embedding is accelerated due to a change in a hyper parameter for domain-specific words.

Abstract: A method and system are provided. The method includes obtaining, by a hardware processor, candidate data representing a plurality of candidates. The method further includes calculating, by the hardware processor, for each of the candidates, a temporal next state of a Recurrent Neural Network (RNN) by inputting a corresponding one of the candidates to the RNN at a current state. The method also includes merging, by the hardware processor, the temporal next state for each of the candidates to obtain the temporal next state of the RNN.

Abstract: Methods and a system are provided for estimating automatic speech recognition (ASR) accuracy. A method includes obtaining transcriptions of utterances in a conversation over two channels. The method further includes sorting the transcriptions along a time axis using a forced alignment. The method also includes training a language model with the sorted transcriptions. The method additionally includes performing ASR for utterances in a conversation between a first user and a second user. The second user is a target of ASR accuracy estimation. The method further includes determining whether an ASR result of the second user is consistent or inconsistent with an ASR result of the first user using the trained language model. The method also includes estimating the ASR result of the second user as poor responsive to the ASR result of the second user being as inconsistent with the ASR result of the first user.

Abstract: A method for reducing response time in a speech interface including constructing a partially completed word sequence from a partially received utterance from a speaker received by an audio sensor, modeling a remainder portion using a processor based on a rich predictive model to predict the remainder portion, and responding to the partially completed word sequence and the predicted remainder portion using a natural language vocalization generator with a vocalization, wherein the vocalization is prepared before a complete utterance is received from the speaker and conveyed to the speaker by an audio transducer.

Abstract: Methods and systems for language processing includes initializing a word embedding matrix based on pre-determined word classes, such that matrix entries associated with a class of which a word is a member are initialized to a non-zero value and other entries are initialized to zero. A neural network is trained based on the initialized word embedding matrix to generate a neural network language model. A language processing task is performed using the neural network language model.

Abstract: Methods and a system are provided for generating labeled data. A method includes encoding, by a processor-based encoder, a first labeled data into an encoded representation of the first labeled data. The method further includes modifying the encoded representation into a modified representation by adding a perturbation to the encoded representation. The method additionally includes decoding, by a processor-based decoder, the modified representation into a second labeled data.

Abstract: A method is provided for training a Deep Neural Network (DNN) for acoustic modeling in speech recognition. The method includes reading central frames and side frames as input frames from a memory. The side frames are preceding side frames preceding the central frames and/or succeeding side frames succeeding the central frames. The method further includes executing pre-training for only the central frames or both the central frames and the side frames and fine-tuning for the central frames and the side frames so as to emphasize connections between acoustic features in the central frames and units of the bottom layer in hidden layer of the DNN.

Abstract: Method of selecting training text for language model, and method of training language model using the training text, and computer and computer program for executing the methods. The present invention provides for selecting training text for a language model that includes: generating a template for selecting training text from a corpus in a first domain according to generation techniques of: (i) replacing one or more words in a word string selected from the corpus in the first domain with a special symbol representing any word or word string, and adopting the word string after replacement as a template for selecting the training text; and/or (ii) adopting the word string selected from the corpus in the first domain as the template for selecting the training text; and selecting text covered by the template as the training text from a corpus in a second domain different from the first domain.

Abstract: Methods and systems for training a neural network include identifying weights in a neural network between a final hidden neuron layer and an output neuron layer that correspond to state matches between a neuron of the final hidden neuron layer and a respective neuron of the output neuron layer. The identified weights are initialized to a predetermined non-zero value and initializing other weights between the final hidden neuron layer and the output neuron layer to zero. The neural network is trained based on a training corpus after initialization.

Abstract: A method for reducing response time in a speech interface including constructing a partially completed word sequence from a partially received utterance from a speaker received by an audio sensor, modeling a remainder portion using a processor based on a rich predictive model to predict the remainder portion, and responding to the partially completed word sequence and the predicted remainder portion using a natural language vocalization generator with a vocalization, wherein the vocalization is prepared before a complete utterance is received from the speaker and conveyed to the speaker by an audio transducer.

Abstract: Embodiments include methods and systems for improving an acoustic model. Aspects include acquiring a first standard deviation value by calculating standard deviation of a feature from first training data and acquiring a second standard deviation value by calculating standard deviation of a feature from second training data acquired in a different environment from an environment of the first training data. Aspects also include creating a feature adapted to an environment where the first training data is recorded, by multiplying the feature acquired from the second training data by a ratio obtained by dividing the first standard deviation value by the second standard deviation value. Aspects further include reconstructing an acoustic model constructed using training data acquired in the same environment as the environment of the first training data using the feature adapted to the environment where the first training data is recorded.

Abstract: Embodiments include methods and systems for improving an acoustic model. Aspects include acquiring a first standard deviation value by calculating standard deviation of a feature from first training data and acquiring a second standard deviation value by calculating standard deviation of a feature from second training data acquired in a different environment from an environment of the first training data. Aspects also include creating a feature adapted to an environment where the first training data is recorded, by multiplying the feature acquired from the second training data by a ratio obtained by dividing the first standard deviation value by the second standard deviation value. Aspects further include reconstructing an acoustic model constructed using training data acquired in the same environment as the environment of the first training data using the feature adapted to the environment where the first training data is recorded.