An embodiment of the present invention provides a speech recognition engine that utilizes portable voice profiles for converting recorded speech to text. Each portable voice profile includes speaker-dependent data, and is configured to be accessible to a plurality of speech recognition engines through a common interface. A voice profile manager receives the portable voice profiles from other users who have agreed to share their voice profiles. The speech recognition engine includes speaker identification logic to dynamically select a particular portable voice profile, in real-time, from a group of portable voice profiles. The speaker-dependent data included with the portable voice profile enhances the accuracy with which speech recognition engines recognize spoken words in recorded speech from a speaker associated with a portable voice profile.

Method And System For Sharing Portable Voice Profiles

PROBLEM TO BE SOLVED: To provide an imaging apparatus capable of properly eliminating the noise component of a mechanism sound included in a sound signal by properly utilizing a noise spectrum at photographing, and a sound recording method and program. SOLUTION: A short time power calculation section 65 calculates the power of a signal of only a motor sound picked up from a reference input section 64 (reference microphone) at the time of photographing and discriminates production timing of the motor sound from the power of the motor sound to carry out noise elimination processing (subtract processing). When the imaging apparatus is provided with a function of amplifying the input sound from a sound input section 51 (main microphone) and adjusting its level by an amplifier 61, a coefficient in accordance with the level of the input sound is multiplied with the noise spectrum (Xv) stored in a storage section 54 and the noise spectrum (Xv) after the coefficient multiplication is subtracted from the spectrum (Ia) of the sound signal. Thus, the noise spectrum is properly used at the time of photographing so that the noise component of the mechanism sound included in the sound signal can properly be eliminated. COPYRIGHT: (C)2007,JPO&INPIT

Imaging apparatus, and sound recording method and program

A system for controlling a human-controlled proxy robot surrogate is presented. The system includes a plurality of motion capture sensors for monitoring and capturing all movements of a human handler such that each change in joint angle, body posture or position; wherein the motion capture sensors are similar in operation to sensors utilized in motion picture animation, suitably modified to track critical handler movements in near real time. A plurality of controls attached to the proxy robot surrogate is also presented that relays the monitored and captured movements of the human handler as “follow me” data to the proxy robot surrogate in which the plurality of controls are configured such that the proxy robot surrogate emulates the movements of the human handler.

Proxy Robots and Remote Environment Simulator for Their Human Handlers

The invention relates to a speech interactive training system and a speech interactive training method. The system comprises a user selection module, a speech interactive training module, a user feedback module, a speech evaluation module and a result feedback module, wherein the user selection module is used for acquiring training contents selected by a user; the speech interactive training module is used for displaying the training contents to the user in a multimode guiding mode to guide the user to perform a speech training; the user feedback module is used for collecting a fed-back speech and a lip video corresponding to the speech; the speech evaluation module is used for receiving the speech fed back by the user and the lip video corresponding to the speech, and automatically evaluating the speech training of the user and giving an evaluation result; and the result feedback module is used for feeding the evaluation result back to the user so that the user can correct and adjustthe speech training. The speech interactive training system is used for automatically evaluating the speech training of the user, giving the evaluation result and feeding the evaluation result back to the user, and then the user finds out the level of the personal speech training according to the evaluation result and corrects and adjusts the personal speech training to further improve the speechlevel, so the rehabitation training effect of a speech impediment is greatly enhanced.

Speech interactive training system and speech interactive training method

An action force being inputted to an object dynamic model (2) is determined such that the motion state amount (object model speed) of the object dynamic model (2) follows up a target motion state amount based on the movement schedule of the object, manipulated variable of the motion state amount (object model speed) of the object dynamic model (2) is determined such that the difference between the actual object position and the target object position approaches 0, the action force and manipulated variable thus determined are then inputted to the object dynamic model (2) thus determining the target object position sequentially. Furthermore, a target object reaction from the object to a robot is determined from the action force thus determined. Consequently, the robot performs a work for moving the object while ensuring stability of the robot by determining the target motion of the object and the target action force between the object and the robot while reducing deviation of actual motion state from the motion state of the object on the object dynamic model.

Controller of leg type moving robot

PROBLEM TO BE SOLVED: To provide a speech recognition device recognizing vowels and consonants according to myoelectric signals. SOLUTION: The speech recognition device 10 is equipped with: a myoelectric signal detecting section 101 which detects the myoelectric signals generated during an utterance action from a plurality of regions; an LPC analysis section 102 which separates and recognizes the respective detected myoelectric signals to spectrum envelope portions and fine change portions based on linear prediction coefficient analyses; a feature extraction section 103 which calculates the linear prediction coefficient cepstrum according to the separated and recognized spectrum envelope portions and calculates the myoelectric signal featured values by each of the channels corresponding to the regions based on the results of the calculation; a likelihood calculating section 106 which receives the myoelectric signal featured values by each of the calculated channels as input vectors and calculates the likelihood based on the hidden Markov model; and a speech recognition section 107 which specifies the speech corresponding to the utterance action based on the calculated likelihood. COPYRIGHT: (C)2005,JPO&NCIPI

Speech recognition device

PROBLEM TO BE SOLVED: To provide a communication terminal device capable of smoothly transmitting gesture to the other side. SOLUTION: The communication terminal device 1a is provided with sensors 80a to 80d which acquire action information for specifying the gesture of a user, a radio part 72 which transmits the acquired action information to another communication terminal device 1b and receives action information for specifying gesture of a user who uses the other communication terminal device 1b from the other communication terminal device 1b, a joint angle driving part 68 which outputs an instruction signal for making a gesture robot 30a which performs a action according to the gesture of the user who uses the other communication terminal device 1b perform the action based on the received action information, first magnetic materials 84a to 84f and a second magnetic material 98 which attachably/detachably load the gesture robot 30a to a user who uses the communication terminal device 1a. COPYRIGHT: (C)2005,JPO&NCIPI

Communication terminal device, action information exchange system, and action information exchange method

The myo-electric information of the user is transmitted to the data center, learning with a high computation load for generating recognition parameters required for recognizing the utterance content based on the myo-electric information is performed, not by the mobile communication terminal, but by the data center. By this, compared with the case when learning is performed in the mobile communication terminal which has stricter limitations on processing speed and memory than a stationary type computer, efficient learning in a shorter time becomes possible. Since the recognition parameters acquired at the data center are transmitted to the mobile communication terminal, hereafter utterance content recognition can be appropriately performed by the recognition parameters matching the user only by the mobile communication terminal, without the mobile communication terminal transmitting/receiving information with a learning device.

Learning device, mobile communication terminal, information recognition system, and learning method

A robot remote manipulation system is provided, including a bipedal walking robot and a remote manipulation device for remotely manipulating the bipedal walking robot. The robot is connected to the remote manipulation device via a communication network and controlled by controlling data from the remote manipulation device. In the system, the remote manipulation device comprises a pair of bilateral mechanical rotating elements providing a quantity of motion for each bilateral leg of the bipedal walking robot; and a controlling data transmitter for transmitting controlling data corresponding to the quantities of motion to the bipedal walking robot. The bipedal walking robot comprises a controlling data receiver for receiving the controlling data transmitted from the remote manipulation device; and a leg motion controller for processing the received controlling data and causing the bilateral legs to move forward or backward.

Robot remote manipulation system and remote manipulation device

This voice detection device is composed of a myoelectric signal acquisition part for acquiring, from a plurality of regions, myoelectric signals generated at the time of a vocalization operation, a parameter calculation part for calculating, as parameters, the fluctuations of the acquired myoelectric signals relative to a predetermined value in every channel corresponding to one of the plurality of regions, a vowel vocalization recognition part for specifying the vowel vocalization operation timing at the time of the vocalization operation, based on the fluctuations of the calculated parameters, and a vowel specification part for specifying a vowel corresponding to the vocalization operation, based on the fluctuation condition of the parameters before and after the specified vocalization operation timing in every channel.

Sugimura Toshiaki

Papers

Speech recognition device

Communication terminal device, action information exchange system, and action information exchange method

Learning device, mobile communication terminal, information recognition system, and learning method

Robot remote manipulation system and remote manipulation device

Voice detection device