Abstract: An information processing apparatus includes a memory; and a processor coupled to the memory and the processor configured to: generate phoneme string information in which a plurality of phonemes included in voice information is arranged in time series, based on a recognition result of the phonemes for the voice information; and learn parameters of a network such that when the phoneme string information is input to the network, output information that is output from the network approaches correct answer information that indicates whether a predetermined conversation situation is included in the voice information that corresponds to the phoneme string information.

Abstract: A non-transitory computer-readable recording medium having stored an evaluation support program that causes a computer to perform a process, the process includes acquiring load information that indicates a load of a subject; acquiring information of an active window of an information processing apparatus on which the subject is working, classifying a type of work, based on the acquired information of the active window, and storing a classification result of the classifying of the type of work, the acquired load information, and time information that indicates acquisition time of the load information in association with each other as information related to a work evaluation of the subject.

Abstract: An apparatus for voice recognition transforms voice information into a phoneme sequence expressed by characters of individual phonemes corresponding to feature parameters of the voice information, determines, based on a first likelihood and a second likelihood, whether or not collation succeeds, executes a matching operation that includes associating, based on a collation result, individual phonemes of the phoneme sequence of the voice information at a time of a failure of collation and individual phonemes of a phoneme sequence of previous voice information with each other, and executes a determination operation that includes determining, based on a result of the association, whether or not the phoneme sequence of the voice information is based on repetitive vocalization.

Abstract: An information processing apparatus includes a memory; and a processor coupled to the memory and the processor configured to: generate phoneme string information in which a plurality of phonemes included in voice information is arranged in time series, based on a recognition result of the phonemes for the voice information; and learn parameters of a network such that when the phoneme string information is input to the network, output information that is output from the network approaches correct answer information that indicates whether a predetermined conversation situation is included in the voice information that corresponds to the phoneme string information.

Abstract: A voice recognition device extracts, from a first voice signal of a user, a first string of phonemes included in the first voice signal, extracts, from a second voice signal of the user, a second string of phonemes included in the second voice signal, extracts a string of common phonemes from the first string and the second string, calculates, for each of a plurality of registered keywords, a degree of similarity between a string of phonemes corresponding to the keyword and the string of common phonemes, and selects, among the plurality of keywords, a prescribed number of keywords based on the degree of similarity for each keyword.

Abstract: A nuisance call detection device that determines whether a telephone call in which a connection is established between a first telephone and a second telephone is a nuisance call and detects the nuisance call, analyzes a voice signal for one telephone call to calculate a plurality of pieces of determination information used for determination as to whether the telephone call is a nuisance call, simultaneously calculates information on a length of a conversation for the one telephone call, further calculates each weighting factor of the plurality of pieces of determination information in accordance with a length of the conversation, then calculates a nuisance call evaluation value based on a value obtained by multiplying each of the plurality of pieces of determination information by the weighting factor, and determines based on the nuisance call evaluation value whether the telephone call is a nuisance call.

Abstract: A state estimating apparatus includes: a spectrum calculating unit which calculates a power spectrum for each of a plurality of frequencies on a frame-by-frame basis from a voice signal containing voice of a first speaker and voice of a second speaker transmitted over a telephone line; a band power calculating unit which calculates power of a non-transmission band on a frame-by-frame basis, based on the power spectra of frequencies contained in the non-transmission band among the plurality of frequencies; a transmitted-voice judging unit which determines that any frame whose power in the non-transmission band is greater than a threshold value indicating the presence of voice carries the voice of the first speaker; and a state judging unit which judges whether the state of mind of the first speaker is normal or abnormal, based on the frame judged to carry the voice of the first speaker.

Abstract: A keyword detector includes a processor configured to calculate a feature vector for each frame from a speech signal, input the feature vector for each frame to a DNN to calculate a first output probability for each triphone according to a sequence of phonemes contained in a predetermined keyword and a second output probability for each monophone, for each of at least one state of an HMM, calculate a first likelihood representing the probability that the predetermined keyword is uttered in the speech signal by applying the first output probability to the HMM, calculate a second likelihood for the most probable phoneme string in the speech signal by applying the second output probability to the HMM, and determine whether the keyword is to be detected on the basis of the first likelihood and the second likelihood.

Abstract: A voice recognition device extracts, from a first voice signal of a user, a first string of phonemes included in the first voice signal, extracts, from a second voice signal of the user, a second string of phonemes included in the second voice signal, extracts a string of common phonemes from the first string and the second string, calculates, for each of a plurality of registered keywords, a degree of similarity between a string of phonemes corresponding to the keyword and the string of common phonemes, and selects, among the plurality of keywords, a prescribed number of keywords based on the degree of similarity for each keyword.

Abstract: An apparatus for voice recognition transforms voice information into a phoneme sequence expressed by characters of individual phonemes corresponding to feature parameters of the voice information, determines, based on a first likelihood and a second likelihood, whether or not collation succeeds, executes a matching operation that includes associating, based on a collation result, individual phonemes of the phoneme sequence of the voice information at a time of a failure of collation and individual phonemes of a phoneme sequence of previous voice information with each other, and executes a determination operation that includes determining, based on a result of the association, whether or not the phoneme sequence of the voice information is based on repetitive vocalization.

Abstract: A keyword detector includes a processor configured to calculate a feature vector for each frame from a speech signal, input the feature vector for each frame to a DNN to calculate a first output probability for each triphone according to a sequence of phonemes contained in a predetermined keyword and a second output probability for each monophone, for each of at least one state of an HMM, calculate a first likelihood representing the probability that the predetermined keyword is uttered in the speech signal by applying the first output probability to the HMM, calculate a second likelihood for the most probable phoneme string in the speech signal by applying the second output probability to the HMM, and determine whether the keyword is to be detected on the basis of the first likelihood and the second likelihood.

Abstract: A nuisance call detection device that determines whether a telephone call in which a connection is established between a first telephone and a second telephone is a nuisance call and detects the nuisance call, analyzes a voice signal for one telephone call to calculate a plurality of pieces of determination information used for determination as to whether the telephone call is a nuisance call, simultaneously calculates information on a length of a conversation for the one telephone call, further calculates each weighting factor of the plurality of pieces of determination information in accordance with a length of the conversation, then calculates a nuisance call evaluation value based on a value obtained by multiplying each of the plurality of pieces of determination information by the weighting factor, and determines based on the nuisance call evaluation value whether the telephone call is a nuisance call.

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: 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: A noise estimation apparatus includes a correlation calculator configured to calculate a correlation value of a spectrum between a plurality of frames in sound information obtained using one or more microphones, a power calculator configured to calculate a power value indicating a sound level of one target frame among the plurality of frames, an update determiner configured to determine an update degree indicating a degree to which the sound information of the target frame is to be reflected in a noise model stored in a storage, or determine whether or not the noise model is to be updated to another noise model, based on the power value of the target frame and the correlation value, and an updater configured to generate the other noise model based on a determined result, the sound information of the target frame, and the noise model.

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: 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: There is provided a noise suppressing device, for suppressing a noise component contained in a sound, including: at least two sound receiving parts receiving sounds from a plurality of directions containing a sound from a direction of a given sound source and converting the sounds to digital sound signals in a time domain, respectively; an estimating part acquiring both direction information on a direction of the given sound source and distance information on a distance from the given sound source based upon the digital sound signals converted by the sound receiving parts, and estimating a component value of a noise component contained in the signal by use of the direction information and the distance information; and a controlling part acquiring a control value of a suppression amount for controlling a range of a direction of the digital sound signals.

Abstract: A sound determination apparatus receives acoustic signals by a plurality of sound receiving units, and generates frames having a predetermined time length. The sound determination apparatus performs FFT on the acoustic signals in frame units, and converts the acoustic signals to a phase spectrum and amplitude spectrum, which are signals on a frequency axis, then calculates the difference at each frequency between the respective acoustic signals as a phase difference, and selects frequencies to be the target of processing. The sound determination apparatus calculates the percentage of frequencies at which the absolute values of the phase differences of the selected frequencies are equal to or greater than a first threshold value, and determines that the acoustic signal coming from the nearest sound source is included in the frame when the calculated percentage is equal to or less than a second threshold value.

Abstract: A state detecting apparatus includes: a processor to execute acquiring utterance data related to uttered speech, computing a plurality of statistical quantities for feature parameters regarding features of the utterance data, creating, on the basis of the plurality of statistical quantities regarding the utterance data and another plurality of statistical quantities regarding reference utterance data based on other uttered speech, pseudo-utterance data having at least one statistical quantity equal to a statistical quantity in the other plurality of statistical quantities, computing a plurality of statistical quantities for synthetic utterance data synthesized on the basis of the pseudo-utterance data and the utterance data, and determining, on the basis of a comparison between statistical quantities of the synthetic utterance data and statistical quantities of the reference utterance data, whether the speaker who produced the uttered speech is in a first state or a second state; and a memory.