Abstract: An information processing system according to the present invention includes a plurality of information processing apparatuses connected by a multihop network, and each of the information processing apparatuses includes a first communication unit configured to transmit and receive a first packet including information about a destination address and a second communication unit configured to communicate at a frequency higher than that of the first communication unit and transmit and receive a second packet including a payload, and in each of the information processing apparatuses, when the first communication unit receives the first packet, the second communication unit is woken up, so that efficient data transfer can be achieved while the power consumption is saved.

Abstract: It is provided a computer comprising a nonvolatile memory for storing data, a control processor for controlling the saving of data into the nonvolatile memory, and a battery for supplying power to the computer in case of a failure of an external power supply, wherein the control processor checks a charge amount stored in the battery, calculates an amount of data which can be saved in the nonvolatile memory by the battery in case of a failure of the external power supply based on the checked charge amount, and saves data excluding the amount of data that can be saved, out of data which should be saved into the nonvolatile memory, into the nonvolatile memory in advance.

Abstract: An information processing system according to the present invention includes a plurality of information processing apparatuses connected by a multihop network, and each of the information processing apparatuses includes a first communication unit configured to transmit and receive a first packet including information about a destination address and a second communication unit configured to communicate at a frequency higher than that of the first communication unit and transmit and receive a second packet including a payload, and in each of the information processing apparatuses, when the first communication unit receives the first packet, the second communication unit is woken up, so that efficient data transfer can be achieved while the power consumption is saved.

Abstract: A synchronization signal generator of a decoder adds a predetermined value to a time value, and outputs a synchronization signal when the added time value coincides with a reproducing time value in a frame.

Abstract: There is provided a sound collecting device, including: an orientation direction forming section that forms an orientation direction of a microphone array; and a control section that, when a characteristic in a frequency band of a synthesized signal obtained by synthesizing the acoustic signals corresponds to a characteristic of an acoustic signal corresponding to a sound other than a target sound, controls the orientation direction forming section such that an orientation direction that is a direction that is different than an orientation direction of the microphone array at a present point in time is formed, and, when the characteristic in the frequency band of the synthesized signal does not correspond to a characteristic of an acoustic signal corresponding to a sound other than the target sound, controls the orientation direction forming section such that the orientation direction of the microphone array is maintained.

Abstract: An encoding apparatus 2 includes: an encoding unit 21 that encodes and multiplexes a video signal V2, an audio signal A2 and a data signal D2 to be given to the encoding apparatus 2; a timer 22 that outputs time information T2; a timer adjusting unit 23 that adjusts the timer 22 so that the time information T2 and time information T1 within a multiplexed stream S1 outputted from an encoding apparatus 1 are synchronized with each other; and a multiplexing unit 24 that multiplexes the multiplexed stream S1, an encoded stream S21 outputted from the encoding unit 21 and the time information T2, and outputs the resultant multiplexed stream and information as output of the encoding apparatus 2. Third and subsequent encoding apparatus are configured to have the same configuration as that of the encoding apparatus 2.

Abstract: The data processor with access protection which has higher reliability during data transfer control according to a transfer condition set by a CPU. The data processor has: a CPU; a memory management section operable to control data transfer by the CPU; and a transfer controller operable to control the data transfer. The transfer controller holds identification information which the memory management section uses for access protection. When producing an address for transfer according to the setting of the CPU, the transfer controller starts data transfer on the condition that the identification information corresponding to the address for transfer matches the identification information of the CPU at the setting of a transfer condition, etc.

Abstract: An echo canceller includes an initial delay estimator producing a reference signal containing an identified frequency component masked by means of frequency masking effect to output the reference signal on an echo path, and detecting, from a sending voice signal, an echo component of the identified frequency component derived from the reference signal. The estimator finds an initial delay amount on the echo path on the basis of a forming timing of the reference signal containing the identified frequency component and a detection timing of the echo component in the identified frequency component in the sending voice signal. The echo canceller includes a delay circuit for delaying a received voice signal by the initial delay amount on the echo path obtained from the estimator to supply a resulting signal to a pseudo echo generator.

Abstract: A synchronization signal generator of a decoder adds a predetermined value to a time value, and outputs a synchronization signal when the added time value coincides with a reproducing time value in a frame.

Abstract: A telephone with a speaker and a microphone uses the speaker both as an electric-to-acoustic transducer, to reproduce the far end party's voice, and an acoustic-to-electric transducer, to capture external sound in the vicinity of the near end party. From the captured sound signal and the signal picked up by the microphone, a decision unit decides whether or not the external sound is being blocked. When the external sound is blocked, a volume control unit increases the volume of the far end party's voice as reproduced by the speaker. The near end party can accordingly turn up the speaker volume by moving the part of the telephone including the speaker close to the user's ear to block external sound, without having to manipulate any manual controls.

Abstract: In an echo canceller, a specific frequency component eliminator eliminates a specific frequency component of a specific frequency from a receiver signal to output a resultant signal to an echo path. A specific frequency component detector detects a frequency component of the same frequency as the specific frequency eliminated by the specific frequency component eliminator from the transmitter signal. A noise calculator finds noise power on the basis of power of the specific frequency component detected by the specific frequency component detector, and finds total power including noise and an echo component on the basis of power of a frequency component including the echo component. A control parameter calculator uses the noise power and the total power found by the noise calculator to find a control parameter of the echo canceller.

Abstract: In an echo canceller including an ADF and a cancellation adder, an ACANC calculator finds an average value of the amount of echo cancellation for a predetermined period to update the average value of the amount of echo cancellation, and detects a convergence state of the ADF to find the amount of echo cancellation in the convergence state of the ADF. A coefficient selector, when receiving a detection signal of the convergence state from the ACANC calculator, finds a power ratio of the sum total of power of all tap coefficients and the partial total of power of tap coefficients not to be assigned to taps to determine the effective number of taps useful for suppressing an echo component on the basis of the amount of echo cancellation and the power ratio. The ADF updates only the effective number of taps to disable the remaining taps.

Abstract: An encoding apparatus 2 includes: an encoding unit 21 that encodes and multiplexes a video signal V2, an audio signal A2 and a data signal D2 to be given to the encoding apparatus 2; a timer 22 that outputs time information T2; a timer adjusting unit 23 that adjusts the timer 22 so that the time information T2 and time information T1 within a multiplexed stream S1 outputted from an encoding apparatus 1 are synchronized with each other; and a multiplexing unit 24 that multiplexes the multiplexed stream S1, an encoded stream S21 outputted from the encoding unit 21 and the time information T2, and outputs the resultant multiplexed stream and information as output of the encoding apparatus 2. Third and subsequent encoding apparatus are configured to have the same configuration as that of the encoding apparatus 2.

Abstract: An echo canceller, in which an adder adds a pseudo echo generated by an adaptive filter to an echo, so that the echo and the pseudo echo are canceled to each other, includes: a subband filter (8, 14) and a sampling converter (9, 15) for downsampling an audio signal; a selector switch (10, 16) which appropriately switches between a side performing the downsampling and another side performing fullband echo cancellation; and a tap arrangement controller (19) which optimizes a tap arrangement by using coefficients which are updated using a downsampled signal by the adaptive filter (11), the tap arrangement controller outputting the optimized tap arrangement to the adaptive filter (11) at time of the fullband echo cancellation.

Abstract: An echo canceler uses an adaptive filter to remove an echo of a far-end input signal from a near-end input signal. Filter coefficients are calculated and updated while the far-end signal is active. While the far-end signal is silent, substitute filter coefficients are similarly calculated to be swapped in as initial values for the filter coefficients when the far-end signal changes from silent to active. The substitute filter coefficients are generated from a simulated far-end signal and a simulated near-end signal obtained by combining corresponding samples in different intervals of the far-end and near-end input signals. To facilitate convergence of the substitute filter coefficients, use of the simulated far-end and near-end signals is disabled when the corresponding samples cancel out.

Abstract: An information processing device having a function for efficiently debugging a parallel processing program by controlling snoop operation is provided. The information processing device is so configured that the following is implemented: the setting for receiving a snoop request from a central processing unit can be set at a snoop controller that controls snoop operation; and as the result of reception of a snoop request, a debug controller can stop multiple central processing units.

Abstract: An uncomfortable feeling due to lack of the speech of a far end speaker and a convergence signal voice (sound) is prevented, and the echo of the far end speaker is removed. A pseudo signal generating circuit has a phase inverter (10) for inverting the phase of a signal, a pseudo hybrid circuit (11) for generating a pseudo echo having the same or substantially the same characteristic as that of the true signal, and a pseudo telephone set (5-2). By generating a pseudo echo having the phase opposite to that of the true signal and having the same or substantially the same characteristics as that of the true signal, the uncomfortable signals (echo, etc.) are cancelled out. When a telephone set is connected to a caller or when an extension transfer occurs, an adaptive filter convergence signal is generated for a predetermined constant time only during “no-voice (sound) (silence)” of the far end speaker's signal.

Abstract: An echo canceller is capable of reducing echo feeling in a conversation even if a tone signal is input. The echo canceller of this invention judges whether a far-end input signal is a calling control tone signal or not, and if that is the case, resets a coefficient of an adaptive filter. An echo canceller of another aspect judges whether a far-end input signal is a predetermined kind of a tone signal or not, and stops the update of the coefficient of the adaptive filter if the result is affirmative and an amount of echo cancellation is not larger than a threshold value. Further, when the tone signal ends, the echo canceller compares amounts of echo cancellation before and after the end of the tone signal, in accordance with a result of the comparison, restart the update of the coefficient of the adaptive filter after resetting the coefficient of the adaptive filter, or restarts the update of the coefficient of the adaptive filter without resetting the coefficient of the adaptive filter.

Abstract: An echo canceller which can respond to a sudden change in echo characteristics in real time and does not require an alteration outside the echo canceller, includes a smoothed sending-speech signal calculation means for calculating a smoothed sending-speech signal from the sending-speech signal, the smoothed sending-speech signal being obtained by smoothing the sending-speech signal; a smoothed receiving-speech signal calculation means for calculating a smoothed receiving-speech signal from the receiving-speech signal, the smoothed receiving-speech signal being obtained by smoothing the receiving-speech signal; a delay time information generation means for obtaining delay time information reflecting delay characteristics of an echo path in accordance with a correlation between the smoothed sending-speech signal and the smoothed receiving-speech signal; and an update information generation means for obtaining update information indicating execution or suspension of updating of the tap coefficients of the adapti

Abstract: An echo canceler uses an adaptive filter to remove an echo of an incoming far-end signal from an outgoing near-end signal in the presence of ambient noise, updating the filter coefficients while the far-end signal is active and the near-end signal is inactive. The near-end and far-end signals are also sampled to obtain data vectors, which are averaged to generate simulated near-end and far-end signals, from which substitute filter coefficients are calculated and updated while the far-end signal is silent, the substitute filter coefficients being used when the far-end signal becomes active again. The number of data vectors averaged is varied according to the ratio of echo power to ambient near-end noise power, or according to the echo attenuation ratio, thereby speeding up convergence of the filter coefficients. Data sampled while the near-end and far-end signals are both active are excluded from the updating process.