Kabushiki Kaisha Toshiba

Masami Akamine

Kimio Miseki

Masahiro Oshikiri

Takehiko Kagoshima

Tadashi Amada

Takami Yoshida

Takashi Masuko

Hideyuki Ueno

Yuka Kobayashi

Kenji Iwata

Vincent Ping Leung Wan

Javier Latorre-Martinez

Kean Kheong Chin

Mark John Francis Gales

Hidetaka Yoshikawa

Yuji Okuda

Javier Latorre

Katherine Mary Knill

Akinobu Yamashita

Koichi Ito

Hiroshi Iwasaki

Hiroshi Fujimura

Masahiro Morita

Koichi Yamamoto

Hiroshi Kanazawa

Yuji Umemoto

Ryosuke Koshiba

Shigenobu Seto

Katsumi Tsuchiya

Koki Otsuka

Yoshinori Shiga

Daniel Barreda

Remco Teunen

Yusuke Shinohara

Byung Ha Chung

Yoshihiro Yamazaki

Takaaki Nii

According to one embodiment, a dialogue system includes a setting apparatus and a processing apparatus. The setting apparatus sets in advance a plurality of words that are in impossible combination relationships to each other. The processing apparatus acquires speech of a user, and when a speech recognition result of an object included in the speech includes a word combination included in the plurality of words that are in impossible combination relationships to each other, output a notification to the user that processing of the object cannot be carried out.

Dialogue system, dialogue method, and storage medium

According to one embodiment, an interactive system includes following units. The knowledge reference unit refers to a question-answering knowledge based on a result of analyzing an input sentence to acquire a candidate for an answer to the input sentence. The unknown keyword detection unit detects, from the input sentence, an unknown keyword. The related keyword estimation unit acquires, in response to the detection of the unknown keyword, one or more candidates for a related keyword having a meaning close to the unknown keyword from predetermined keywords. The response generation unit generates a response to the input sentence based on the one or more candidates for the related keyword when the unknown keyword is detected.

Interactive system, apparatus, and method

According to an embodiment, a dialogue system includes a satisfaction estimator, a dialogue state estimator, and a behavior determiner. The satisfaction estimator estimates a satisfaction of a user based on a speech input from the user. The dialogue state estimator estimates a dialogue state with the user based on the speech input from the user and the estimated satisfaction of the user. The behavior determiner determines a behavior towards the user based on the estimated dialogue state.

Dialogue system and dialogue method

A text-to-speech method for simulating a plurality of different voice characteristics includes dividing inputted text into a sequence of acoustic units; selecting voice characteristics for the inputted text; converting the sequence of acoustic units to a sequence of speech vectors using an acoustic model having a plurality of model parameters provided in clusters each having at least one sub-cluster and describing probability distributions which relate an acoustic unit to a speech vector; and outputting the sequence of speech vectors as audio with the selected voice characteristics. A parameter of a predetermined type of each probability distribution is expressed as a weighted sum of parameters of the same type using voice characteristic dependent weighting. In converting the sequence of acoustic units to a sequence of speech vectors, the voice characteristic dependent weights for the selected voice characteristics are retrieved for each cluster such that there is one weight per sub-cluster.

Text to speech method and system using voice characteristic dependent weighting

A text-to-speech method configured to output speech having a selected speaker voice and a selected speaker attribute, including: inputting text; dividing the inputted text into a sequence of acoustic units; selecting a speaker for the inputted text; selecting a speaker attribute for the inputted text; converting the sequence of acoustic units to a sequence of speech vectors using an acoustic model; and outputting the sequence of speech vectors as audio with the selected speaker voice and a selected speaker attribute. The acoustic model includes a first set of parameters relating to speaker voice and a second set of parameters relating to speaker attributes, which parameters do not overlap. The selecting a speaker voice includes selecting parameters from the first set of parameters and the selecting the speaker attribute includes selecting the parameters from the second set of parameters.

Text to speech system

According to one embodiment, a speech synthesizer includes an analyzer, a first estimator, a selector, a generator, a second estimator, and a synthesizer. The analyzer analyzes text and extracts a linguistic feature. The first estimator selects a first prosody model adapted to the linguistic feature and estimates prosody information that maximizes a first likelihood representing probability of the selected first prosody model. The selector selects speech units that minimize a cost function determined in accordance with the prosody information. The generator generates a second prosody model that is a model of the prosody information of the speech units. The second estimator estimates prosody information that maximizes a third likelihood calculated on the basis of the first likelihood and a second likelihood representing probability of the second prosody model. The synthesizer generates synthetic speech by concatenating the speech units on the basis of the prosody information estimated by the second estimator.

Speech synthesizer, speech synthesizing method and program product

A speech processing apparatus, including a segmenting unit to divide a fundamental frequency signal of a speech signal corresponding to an input text into pitch segments, based on an alignment between samples of at least one given linguistic level included in the input text and the speech signal. Character strings of the input text are divided into the samples based on each linguistic level. A parameterizing unit generates a parametric representation of the pitch segments using a predetermined invertible operator and generates a group of first parameters in correspondence with each linguistic level. A descriptor generating unit generates, for each linguistic level, a descriptor that includes a set of features describing each sample in the input text and a model learning unit classifies the first parameters of each linguistic level of all speech signals in a memory into clusters based on the descriptor corresponding to the linguistic level.

Speech processing apparatus, method, and computer program product for synthesizing speech

A spectrum calculating unit calculates, for each of the frames, a spectrum by performing a frequency analysis on an acoustic signal. An estimating unit estimates a noise spectrum. An energy calculating unit calculates an energy characteristic amount. An entropy calculating unit calculates a normalized spectral entropy value. A generating unit generates a characteristic vector based on the energy characteristic amounts and the normalized spectral entropy values that have been calculated for a plurality of frames. A likelihood calculating unit calculates a speech likelihood value of a target frame that corresponds to the characteristic vector. In a case where the speech likelihood value is larger than a threshold value, a judging unit judges that the target frame is a speech frame.

Apparatus, method, and computer program product for judging speech/non-speech

A noise-environment storing unit stores therein a compensation vector for compensating a feature vector of a speech. A feature-vector extracting unit extracts the feature vector of the speech in each of a plurality of frames. A noise-environment-series estimating unit estimates a noise-environment series based on a feature-vector series and a degree of similarity. A calculating unit obtains a compensation vector corresponding to each noise environment in estimated noise-environment series based on the compensation vector present in the noise-environment storing unit. A compensating unit compensates the extracted feature vector of the speech based on obtained compensation vector.

Feature-vector compensating apparatus, feature-vector compensating method, and computer program product

A transformation-parameter calculating unit calculates a first model parameter indicating a parameter of a speaker model for causing a first likelihood for a clean feature to maximum, and calculates a transformation parameter for causing the first likelihood to maximum. The transformation parameter transforms, for each of the speakers, a distribution of the clean feature corresponding to the identification information of the speaker to a distribution represented by the speaker model of the first model parameter. A model-parameter calculating unit transforms a noisy feature corresponding to identification information for each of speakers by using the transformation parameter, and calculates a second model parameter indicating a parameter of the speaker model for causing a second likelihood for the transformed noisy feature to maximum.

Apparatus for creating speaker model, and computer program product

Normalization parameters are generated at a normalization-parameter generating unit by calculating the mean values and the standard deviations of an initial prosody pattern and a prosody pattern of a training sentence of a speech corpus. Then, the variance range or variance width of the initial prosody pattern is normalized at the prosody-pattern normalizing unit in accordance with the normalization parameters. As a result, a prosody pattern similar to speech of human beings and improved in naturalness can be generated with a small amount of calculation.

Prosody-pattern generating apparatus, speech synthesizing apparatus, and computer program product and method thereof

A speech synthesis method comprises selecting a predetermined formant parameters from formant parameters according to a pitch pattern, phoneme duration, and phoneme symbol string, generating a plurality of sine waves based on formant frequency and formant phase of the formant parameters selected, multiplying the sine waves by windowing functions of the selected formant parameters, respectively, to generate a plurality of formant waveforms, adding the formant waveforms to generate a plurality of pitch waveforms, and superposing the pitch waveforms according to a pitch period to generate a speech signal.

Speech synthesis method and speech synthesizer

A speech encoding method, apparatus and program wherein an input speech signal is divided into a plurality of frames each having a predetermined length, each of the frames is subdivided into a plurality of subframes, a predictive pitch period of a subframe in a to-be-encoded current frame is obtained by using pitch periods of at least two frames of the current frame and past and future frames with respect to the current frame; a pitch period of a subframe in the current frame is obtained by using the predictive pitch period, a relative pitch pattern codebook storing a plurality of relative pitch patterns representing fluctuations in pitch periods of a plurality of subframes is prepared, and a change in pitch period of plural subframes is expressed with one relative pitch pattern selected from the relative pitch pattern codebook.

Speech encoding method, apparatus and program

A speech synthesis method subjects a reference speech signal to windowing to extract a speech pitch wave having a window function of a window length double a pitch period of the reference speech signal from the reference speech signal. A linear prediction coefficient is generated by subjecting the reference speech signal to a linear prediction analysis. The speech pitch wave is subjected to inverse-filtering based on the linear prediction coefficient to produce a residual pitch wave, which is then stored as information of a speech synthesis unit in a voiced period in a storage. Speech using the information of the speech synthesis unit is then synthesized.

Speech synthesis method

A speech synthesis method that generates a speech pitch wave from a reference speech signal by subjecting the reference speech signal to one of Fourier transform and Fourier series expansion to produce a discrete spectrum, that interpolates the discrete spectrum to generate a consecutive spectrum, and that subjects the consecutive spectrum to inverse Fourier transform. A linear prediction coefficient is generated by subjecting the reference speech signal to a linear prediction analysis. The speech pitch wave is subjected to inverse-filtering based on the linear prediction coefficient to produce a residual pitch wave. Information regarding the residual pitch wave is stored as information of a speech synthesis unit in a voice period. A speech is then synthesized using the information of the speech synthesis unit.

Growing community of inventors

Kanagawa, Japan

Masami Akamine