Nippon Telegraph and Telephone Corporation

The University of Tokyo

Noboru Harada

Takehiro Moriya

Yutaka Kamamoto

Ryosuke Sugiura

Masahiro Fukui

Yusuke Hiwasaki

Hirokazu Kameoka

Yuma Koizumi

Shoichiro Saito

Sachiko Kurihara

Takuya Nishimoto

Akira Nakagawa

Takahito Kawanishi

Shigeki Sagayama

Shin Murata

Kazunori Kobayashi

Dai Yang

Kenta Niwa

Yuta Kawachi

Hironobu Chiba

Kazunaga Ikeda

Keisuke Hasegawa

Akio Jin

Kenichi Noguchi

Jouji Nakayama

Masataka Yamaguchi

Hiroshi Sekigawa

Masahiro Yasuda

Kiyoshi Shirayanagi

Kouichi Furukado

Junichi Nakajima

An autocorrelation calculation unitcalculates an autocorrelation R(i) from an input signal. A prediction coefficient calculation unitperforms linear prediction analysis by using a modified autocorrelation R'(i) obtained by multiplying a coefficient w(i) by the autocorrelation R(i). It is assumed here, for each order i of some orders i at least, that the coefficient w(i) corresponding to the order i is in a monotonically increasing relationship with an increase in a value that is negatively correlated with a fundamental frequency of the input signal of the current frame or a past frame.

Linear prediction analysis device, method, program, and storage medium

Provided is an abnormality estimation device capable of appropriately determining normal data appearing less frequently as normal. The abnormality estimation device includes an estimation unit that estimates an anomaly degree of an acoustic signal, by using an abnormality estimation model that is optimized while using a set of normal sounds and is optimized so as to minimize a difference between an anomaly degree of a normal sound appearing more frequently and an anomaly degree of a normal sound appearing less frequently.

Abnormality estimation device, abnormality estimation method, and program

The number of conversational tests required for evaluation of acoustic quality of the ICC system is reduced. An evaluation value converting deviceevaluates the quality of a conversation made across a near-end acoustic areaand a far-end acoustic areainside a vehicle in which a plurality of acoustic areas are predetermined. A voice signal collected by a microphone Mdisposed in the far-end acoustic areais emitted from a speaker Sdisposed in the near-end acoustic area. An objective evaluation value acquisition unitacquires an evaluation value using, as an evaluation target sound, a voice signal obtained by adding a voice signal obtained from a first voice signal emitted from a sound source in the far-end acoustic area, collected by the microphone M, and emitted from the speaker Sto a voice signal obtained from the first voice signal transmitted through a space inside the vehicle and reaching the near-end acoustic area, with the first voice signal being used as a reference sound.

Evaluation apparatus, evaluation method, and program

An anomaly detection technique which realizes high accuracy while reducing cost required for normal model learning is provided. An anomaly detection apparatus includes an anomaly degree estimating unit configured to estimate an anomaly degree indicating a degree of anomaly of anomaly detection target equipment from sound emitted from the anomaly detection target equipment (hereinafter, referred to as anomaly detection target sound) based on association between a first probability distribution indicating distribution of normal sound emitted from one or more pieces of equipment different from the anomaly detection target equipment and normal sound emitted from the anomaly detection target equipment (hereinafter, referred to as normal sound for adaptive learning).

Anomaly detection apparatus, probability distribution learning apparatus, autoencoder learning apparatus, data transformation apparatus, and program

A possible region of encoding results of anomalous samples is limited. An encoder storage unitstores an encoder for projecting an input feature value into a latent space in which the latent space is a closed manifold, a normal distribution obtained by learning normal data and an anomalous distribution obtained by learning anomalous data are held on the manifold, and a decoder for reconstructing the output of the encoder. An encoding unitobtains a reconstruction result output by the decoder when a feature value of target data is input to the encoder. An anomaly score calculation unitcalculates an anomaly score of the target data based on distances between the reconstruction result and the normal distribution and distances between the reconstruction result and the anomalous distribution.

Anomaly detection apparatus, anomaly detection method, and program

The number of conversational tests needed for the evaluation of acoustic quality of the ICC system is reduced. An evaluation value conversion deviceevaluates the quality of communication between a near-end acoustic regionand a far-end acoustic regioninside a vehicle for which a plurality of acoustic regions are predetermined. A voice signal picked up by a microphone Mdisposed in the far-end acoustic regionis emitted from a speaker Sdisposed in the near-end acoustic region. An objective evaluation value acquisition unitacquires a first evaluation value by treating a voice signal obtained by combining a voice signal resulting from a first voice signal being emitted from a sound source in a seat belonging to the far-end acoustic region, picked up by the microphone M, and emitted from the speaker Swith a voice signal arriving at a seat belonging to the near-end acoustic regionas a result of the first voice signal being transmitted through the space inside the vehicle as an evaluation target sound, and treating the first voice signal as a reference sound. An evaluation value reuse unitacquires the first evaluation value as an evaluation value of communication between a seat belonging to the near-end acoustic regionand a seat belonging to the far-end acoustic region

Evaluation method, evaluation apparatus, and program

A coding method and a decoding method are provided which can use in combination a predictive coding and decoding method which is a coding and decoding method that can accurately express coefficients which are convertible into linear prediction coefficients with a small code amount and a coding and decoding method that can obtain correctly, by decoding, coefficients which are convertible into linear prediction coefficients of the present frame if a linear prediction coefficient code of the present frame is correctly input to a decoding device. A coding device includes: a predictive coding unit that obtains a first code by coding a differential vector formed of differentials between a vector of coefficients which are convertible into linear prediction coefficients of more than one order of the present frame and a prediction vector containing at least a predicted vector from a past frame, and obtains a quantization differential vector corresponding to the first code; and a non-predictive coding unit that generates a second code by coding a correction vector which is formed of differentials between the vector of the coefficients which are convertible into the linear prediction coefficients of more than one order of the present frame and the quantization differential vector or formed of some of elements of the differentials.

Coding device, decoding device, and method and program thereof

The accuracy of unsupervised anomalous sound detection is improved. An anomalous sound generation unitgenerates a pseudo-anomalous sound by associating a sound obtained from a predetermined device with a probability distribution followed by a sound obtained from a desired device group. A threshold value setting unitsets a threshold value so that all anomaly scores calculated from the pseudo-anomalous sound are determined to be anomalous. A parameter update unituses an obtained normal sound and an obtained anomalous sound, or a threshold value and a value based on the obtained anomalous sound, to update a parameter so to reliably determine an anomalous sound as anomalous and minimize the probability of determining a normal sound as anomalous.

Sound generation apparatus, data generation apparatus, anomaly score calculation apparatus, and program

An autocorrelation calculation unitcalculates an autocorrelation R(i) from an input signal. A prediction coefficient calculation unitperforms linear prediction analysis by using a modified autocorrelation R′(i) obtained by multiplying a coefficient w(i) by the autocorrelation R(i). It is assumed here, for each order i of some orders i at least, that the coefficient w(i) corresponding to the order i is in a monotonically increasing relationship with an increase in a value that is negatively correlated with a fundamental frequency of the input signal of the current frame or a past frame.

To improve accuracy of an evaluation in an acoustic quality evaluation test performed by comparing an evaluation target sound and a reference sound. A data correction apparatuscompares, in a call performed between a near-end terminaland a far-end terminal, an evaluation target sound in which a voice output from the near-end terminalis recorded and a reference sound in which a voice spoken by a call partner using the far-end terminalto correct test data used in a listening test for evaluating acoustic quality of the call. A correction target determination unitdetermines, as a correction target section, a voiced section that does not include the voice of the call partner detected from an acoustic signal representing the reference sound. A correction execution unitupdates the correction target section of the acoustic signal representing the reference sound with a non-voice signal predetermined.

Data correction apparatus, data correction method, and program

Sound source enhancement technology is provided that is capable of improving sound source enhancement capabilities in accordance with settings of usage and applications. A PSD optimization device includes a PSD updating unit that takes a target sound PSD input value {circumflex over ( )}φS(ω, τ), an interference noise PSD input value {circumflex over ( )}φIN(ω, τ), and a background noise PSD input value {circumflex over ( )}φBN(ω, τ) as input, and generates a target sound PSD output value φS(ω, τ), an interference noise PSD output value {circumflex over ( )}φIN(ω, τ), and a background noise PSD output value {circumflex over ( )}φBN(ω, τ), by solving an optimization problem for a cost function relating to a variable uS representing a target sound PSD, a variable uIN representing an interference noise PSD, and a variable uBN representing a background noise PSD. The optimization problem for the cost function is defined using at least one of a constraint relating to a frequency structure of a sound source or a convex cost term relating to the frequency structure of the sound source, a constraint relating to a temporal structure of the sound source or a convex cost term relating to the temporal structure of the sound source, and a constraint relating to a spatial structure of the sound source or a convex cost term relating to the spatial structure of the sound source.

PSD optimization apparatus, PSD optimization method, and program

An input of a first observation signal corresponding to an incoming signal from a first direction is received, an angular rotation operation of the first observation signal is performed to obtain a second observation signal corresponding to an incoming signal from a second direction that is different from the first direction and the second observation signal is added to a set of training data.

Training data extension apparatus, training data extension method, and program

An envelope sequence is provided that can improve approximation accuracy near peaks caused by the pitch period of an audio signal. A periodic-combined-envelope-sequence generation device according to the present invention takes, as an input audio signal, a time-domain audio digital signal in each frame, which is a predetermined time segment, and generates a periodic combined envelope sequence as an envelope sequence. The periodic-combined-envelope-sequence generation device according to the present invention comprises at least a spectral-envelope-sequence calculating part and a periodic-combined-envelope generating part. The spectral-envelope-sequence calculating part calculates a spectral envelope sequence of the input audio signal on the basis of time-domain linear prediction of the input audio signal. The periodic-combined-envelope generating part transforms an amplitude spectral envelope sequence to a periodic combined envelope sequence on the basis of a periodic component of the input audio signal in the frequency domain.

Periodic-combined-envelope-sequence generation device, periodic-combined-envelope-sequence generation method, periodic-combined-envelope-sequence generation program and recording medium

Provided is an acoustic signal processing technique for performing a signal transformation suitable for desired signal processing (e.g., sound source enhancement processing) on a signal, and then performing the desired signal processing on the transformed signal. An acoustic signal processing device performs signal processing M which is a desired target on an input acoustic signal x. The acoustic signal processing device includes a transform unit that performs transform processing P on the acoustic signal x to obtain a first transform coefficient X; a signal processing unit that performs signal processing M corresponding to a desired target on the first transform coefficient X to obtain a second transform coefficient {circumflex over (&#x2003;)}S; and an inverse transform unit that performs inverse transform processing Pon the second transform coefficient {circumflex over (&#x2003;)}S to obtain an acoustic signal {circumflex over (&#x2003;)}s subjected to signal processing which is a desired target, wherein the transform processing P, the inverse transform processing P, and the signal processing M are simultaneously optimized.

Acoustic signal processing apparatus, method and program for the same

Provided is sound source enhancement technology that is capable of improving sound source enhancement capabilities in a configuration using a beamformer to suppress interference noise. A PSD optimization device includes a PSD updating unit that takes a target sound PSD input value, an interference noise PSD input value, and a background noise PSD input value as input, and generates a target sound PSD output value, an interference noise PSD output value, and a background noise PSD output value, by solving an optimization problem for a cost function relating to a variable representing a target sound PSD, a variable representing an interference noise PSD, and a variable representing a background noise PSD. The optimization problem is defined using at least one of a constraint or a convex cost term based on difference in sound source enhancement characteristics of a beamformer in accordance with a degree of inclusion of a target sound, with regard to a PSD of signals beamformed in a target sound direction-of-arrival, and a constraint or a convex cost term based on the difference in sound source enhancement characteristics of the beamformer in a frequency direction, with regard to the PSD of signals beamformed in the target sound direction-of-arrival.

Growing community of inventors

Atsugi, Japan

Noboru Harada