Abstract: A method, computer system, and a computer program product for audio data augmentation are provided. Sets of audio data from different sources may be obtained. A respective normalization factor for at least two sources of the different sources may be calculated. The normalization factors from the at least two sources may be mixed to determine a mixed normalization factor. A first set of the sets may be normalized by using the mixed normalization factor and to obtain training data for training an acoustic model.

Abstract: A system and method for expanding a question and answer (Q&A) database. The method includes obtaining a set of Q&A documents and speech recognition results, each Q&A document in the set having an identifier, and each speech recognition result having an identifier common with the identifier of a relevant Q&A document, and adding one or more repetition parts extracted from the speech recognition results to a corresponding Q&A document in the set to generate an expanded set of Q&A documents for increasing Q&A document extraction accuracy.

Abstract: An approach to identifying alternate soft labels for training a student model may be provided. A teaching model may generate a soft label for a labeled training data. The training data can be an acoustic file for speech or a spoken natural language. A pool of soft labels previously generated by teacher models can be searched at the label level to identify soft labels that are similar to the generated soft label. The similar soft labels can have similar length or sequence at the word phoneme, and/or state level. The identified similar soft labels can be used in conjunction with the generated soft label to train a student model.

Abstract: A technique for data augmentation for speech data is disclosed. Original speech data including a sequence of feature frames is obtained. A partially prolonged copy of the original speech data is generated by inserting one or more new frames into the sequence of the feature frames. The partially prolonged copy is output as augmented speech data for training an acoustic model for training an acoustic model.

Abstract: A technique for estimating phonemes for a word written in a different language is disclosed. A sequence of graphemes of a given word in a source language is received. The sequence of the graphemes in the source language is converted into a sequence of phonemes in the source language. One or more sequences of phonemes in a target language are generated from the sequence of the phonemes in the source language by using a neural network model. One sequence of phonemes in the target language is determined for the given word. Also, technique for estimating graphemes of a word from phonemes in a different language is disclosed.

Abstract: A method, computer program product, and apparatus for adapting a trained neural network having one or more batch normalization layers are provided. The method includes adapting only the one or more batch normalization layers using adaptation data. The method also includes adapting the whole of the neural network having the one or more adapted batch normalization layers, using the adaptation data.

Abstract: A technique for estimating phonemes for a word written in a different language is disclosed. A sequence of graphemes of a given word in a source language is received. The sequence of the graphemes in the source language is converted into a sequence of phonemes in the source language. One or more sequences of phonemes in a target language are generated from the sequence of the phonemes in the source language by using a neural network model. One sequence of phonemes in the target language is determined for the given word. Also, technique for estimating graphemes of a word from phonemes in a different language is disclosed.

Abstract: A technique for detecting a music segment in an audio signal is disclosed. A time window is set for each section in an audio signal. A maximum and a statistic of the audio signal within the time window are calculated. A density index is computed for the section using the maximum and the statistic. The density index is a measure of the statistic relative to the maximum. The section is estimated as a music segment based, at least in part, on a condition with respect to the density index.

Abstract: A computer-implemented method is provided for generating a plurality of templates. The method includes obtaining, by a processor device, a Recurrent Neural Network Language Model (RNNLM) trained using a first set of text data. The method further includes adapting, by the processor device, the RNNLM using a second set of text data by adding a new node corresponding to a class in both an input layer and an output layer of the RNNLM, the class being obtained from the second set of text data. The method also includes generating, by the processor device, the plurality of templates using the adapted RNNLM.

Abstract: A technique for data augmentation for speech data is disclosed. Original speech data including a sequence of feature frames is obtained. A partially prolonged copy of the original speech data is generated by inserting one or more new frames into the sequence of the feature frames. The partially prolonged copy is output as augmented speech data for training an acoustic model for training an acoustic model.

Abstract: A computer-implemented method, computer program product, and system are provided for separating a word in a dictionary. The method includes reading a word from the dictionary as a source word. The method also includes searching the dictionary for another word having a substring with a same surface string and a same reading as the source word. The method additionally includes splitting the another word by the source word to obtain one or more remaining substrings of the another word. The method further includes registering each of the one or more remaining substrings as a new word in the dictionary.

Abstract: A computer-implemented method is provided for generating a plurality of templates. The method includes obtaining, by a processor device, a Recurrent Neural Network Language Model (RNNLM) trained using a first set of text data. The method further includes adapting, by the processor device, the RNNLM using a second set of text data by adding a new node corresponding to a class in both an input layer and an output layer of the RNNLM, the class being obtained from the second set of text data. The method also includes generating, by the processor device, the plurality of templates using the adapted RNNLM.

Abstract: A technique for detecting a music segment in an audio signal is disclosed. A time window is set for each section in an audio signal. A maximum and a statistic of the audio signal within the time window are calculated. A density index is computed for the section using the maximum and the statistic. The density index is a measure of the statistic relative to the maximum. The section is estimated as a music segment based, at least in part, on a condition with respect to the density index.

Abstract: A computer-implemented method, computer program product, and system are provided for separating a word in a dictionary. The method includes reading a word from the dictionary as a source word. The method also includes searching the dictionary for another word having a substring with a same surface string and a same reading as the source word. The method additionally includes splitting the another word by the source word to obtain one or more remaining substrings of the another word. The method further includes registering each of the one or more remaining substrings as a new word in the dictionary.

Abstract: A computer-implemented method, computer program product, and system are provided for separating a word in a dictionary. The method includes reading a word from the dictionary as a source word. The method also includes searching the dictionary for another word having a substring with a same surface string and a same reading as the source word. The method additionally includes splitting the another word by the source word to obtain one or more remaining substrings of the another word. The method further includes registering each of the one or more remaining substrings as a new word in the dictionary.

Abstract: A method for processing a speech signal. The method comprises obtaining a logmel feature of a speech signal. The method further includes one or more processors processing the logmel feature so that the logmel feature is normalized under a constraint that a power level of the logmel feature is kept as originally obtained. The method further includes inputting the processed logmel feature into a speech-to-text system to generate corresponding text data.

Abstract: A system and method for expanding a question and answer (Q&A) database. The method includes obtaining a set of Q&A documents and speech recognition results, each Q&A document in the set having an identifier, and each speech recognition result having an identifier common with the identifier of a relevant Q&A document, and adding one or more repetition parts extracted from the speech recognition results to a corresponding Q&A document in the set to generate an expanded set of Q&A documents for increasing Q&A document extraction accuracy.

Abstract: A computer-implemented method, computer program product, and system are provided for separating a word in a dictionary. The method includes reading a word from the dictionary as a source word. The method also includes searching the dictionary for another word having a substring with a same surface string and a same reading as the source word. The method additionally includes splitting the another word by the source word to obtain one or more remaining substrings of the another word. The method further includes registering each of the one or more remaining substrings as a new word in the dictionary.

Abstract: A system and method for expanding a question and answer (Q&A) database. The method includes preparing a set of Q&A documents and speech recognition results of an agent's utterances in conversations between an agent and a customer, each Q&A document in the set having an identifier, and each speech recognition result having an identifier common with the identifier of a relevant Q&A document, and adding one or more repetition parts extracted from the speech recognition results of the agent's utterances to a corresponding Q&A document in the set.

Abstract: A method, computer program product, and apparatus for adapting a trained neural network having one or more batch normalization layers are provided. The method includes adapting only the one or more batch normalization layers using adaptation data. The method also includes adapting the whole of the neural network having the one or more adapted batch normalization layers, using the adaptation data.