Fujitsu Corporation

Yasushi Yamazaki

Hitoshi Matsuzawa

Takeshi Otani

Tomonori Sato

Tomohiko Taniguchi

Hitoshi Sasaki

Hisanari Kimura

Kazuhiro Watanabe

Yasuji Ota

Yoshihiro Tomita

Nobuyuki Washio

Kaori Endo

Hiroshi Katayama

Nobuyuki Katae

Mutsumi Saito

Chiharu Kawai

A voice synthesizer, which obtains a voice by emphasizing a specific part of a sentence, includes an emphasis degree deciding unit that extracts a word or a collocation to be emphasized from among respective words or respective collocations on the basis of an extracting reference with respect to the each word or the each collocation included in a sentence and deciding an emphasis degree of the extracted word or the extracted collocation, an acoustic processing unit that synthesizes a voice having an emphasis degree which is decided by the emphasis degree deciding unit applied to the word to be emphasized or the collocation to be emphasized, whereby the emphasized part of the word or the collocation can be obtained automatically on the basis of the extracting reference, such as a frequency of appearance and a level of importance of the word or the collocation.

Word or collocation emphasizing voice synthesizer

A speech input device is provided with a microphone which inputs speech, a key entry detector which detects an operation of a key section which serves as a man-machine interface, and a noise eliminator which eliminates a component of an operation sound from the speech that is input into the microphone within a period in which the key entry detector detects the operation.

Speech input device

In an apparatus which estimates characteristics of a surrounding noise only when an input signal is soundless and performs a noise reduction or suppression of the input signal based on the estimated result, a signal noise ratio is estimated from the input signal, and an automatic switch or an automatic adjustment is performed so as to execute a noise reduction only when the signal noise ratio is good, otherwise to avoid the noise reduction or make the noise reduction degree smaller.

Noise suppressing apparatus

The collating device includes the Formant estimation section that estimates feature amount data that represents a time-based change in frequencies of first and second Formants from voice data input through the microphone. The difference computing section calculates a difference between some reference feature amount data and the feature amount data as the feature-amount difference data. This feature-amount difference data is stored in a database. The feature-amount difference data corresponding to voice data input upon collation is collated with the feature-amount difference data registered in the database, and the result is output.

Method of and device for collating data, and computer product

An echo canceler makes it possible to adaptably control a power of speech reception voice and to reduce distortion of a voice output from a speaker. The echo canceler has a basic structure includes a power control section controlling a power for a speech reception voice, a first filtering section removing a particular component from an output of the power control section, an echo canceling section removing echo components of a speech reception voice, which are added to a speech transmission voice, a second filtering section filtering particular components of the speech reception and transmission voices, and a state judgement section judging existences of the speech reception and transmission voices from an output of the second filtering section and controlling a power attenuation amount of the power control section based on the judgement result.

Echo canceler

In an active noise control method of the present invention, a noise detection signal from a noise referring microphone which detects ambient noise is inputted into noise processing units of different types to produce noise silencing signals. A selected noise silencing signal is inputted into a receiver or a speaker together with a reception signal via an adder, so that a regenerated tone from the noise silencing signal silences the ambient noise, When a nonconformance detecting unit judges that the selected noise silencing signal exceeds a predetermined range, the corresponding noise processing unit is judged to be incompatible, and a switch control unit controls a switch to select another one of the noise processing units.

Active noise control method and receiver device

A stereophonic sound processing system for localizing a sound image at desired locations using devices including a headphone and a speaker, includes a processing unit for generating stereophonic sound on the basis of an input signal. The processing unit includes: a plurality of stereophonic filter units each including an FIR filter for processing an input signal; a selecting unit for selecting one of the plurality of stereophonic filter units in accordance with a desired factor; wherein the processing unit controls the selected stereophonic filter unit so as to generate stereophonic sound on the basis of a processed result supplied by the selected stereophonic filter unit.

Stereophonic sound processing system

An apparatus for continuously reproducing plural sound data has a start end/terminal end determination unit for determining the start end/terminal end of the continued respective sound data, a fade-in/fade-out unit for carrying out fade-in process at the start end of plural respective sound data and/or fade-out process at the terminal end of the same, a data output unit for continuously outputting the plural sound data which have been subjected to fade-in process and/or fade-out process, and a reproduction unit for reproducing the outputted plural sound data. In reproducing continuously the plural sound data, no noise is generated at the joint portion of the adjacent sound data.

Sound reproduction method, sound reproduction apparatus, sound data creation method, and sound data creation apparatus

Speech processing coder, decoder and command recognizer

To improve the voice quality of a digital mobile communication system such as a car telephone or portable telephone when outdoor background noises are superimposed on voices. To achieve the above object, a voice decoding apparatus comprises noise superimposed part detecting means for discriminating between a noise part containing only noises and a voice part containing voices from signal encoded at a transmission side, voice decoding means for decoding an encoded signal in the voice part into a waveform signal, noise decoding means for decoding an encoded signal in the noise part into a waveform signal, and noise control means for controlling the frequency characteristic of said noise part by controlling said noise decoding means when said noise superimposed part detecting means judges the noise part.

Voice encoding and voice decoding apparatus

A link information storer stores link connection information and a node information storer stores node label information by sequentially searching links from the beginning of a network when data in a network structure are configured by linking nodes according to the relations among nodes (N), which represent basic units of data. The link information storer stores a previous or next node identifier of each link. When a plurality of links share the same next node, their pieces of link information are stored successively. To indicate that the next nodes are the same, a positive or negative sign is attached to the node numbers showing the previous nodes. Link numbers are uniquely assigned to respective links. A node identifier is made to correspond with a link identifier, which is realized either by assigning as the value of the node identifier the link number of a link connecting the node to the next node, or the maximum or minimum value of the link numbers of links connecting the node to the next nodes, or by attaching a unique node number to a node as with attaching a unique link number to a link.

Network structure storing and retrieval method for a data processor

A dynamic programming or DP matching system for speech recognition. Upon DP matching, cumulative distance is compared with a threshold value at every sampling time point of a speech pattern to thereby restrict the number of DP paths in succeeding matching processes. The number of DP paths remaining at each speech pattern sampling time point is monitored by a monitoring module for altering a threshold value so as to decrease the number of the DP paths. When DP path number becomes excessively large, the threshold is increased to thereby decrease the DP path number. Capacity of a DP data storing memory can be reduced while preventing the matching capability from being lowered.

Growing community of inventors

Kawasaki, Japan

Yasushi Yamazaki