Abstract: The continual engagement support device 600 includes: the performance data storage means 601 that stores performance data including time data indicating the time of performance and sensor data of the time of performance or the surrounding time obtained from the sensors mounted on the portable terminal when the target action is performed via the portable terminal possessed by the target user; the timing detection means 602 that detects the performance timing, which is the timing at which the target user is likely to perform the target action during the predetermined cycle, on the basis of the performance data; and the notification means 603 provides a notification to prompt the performance of the target action at the detected performance timing when the target action is not performed for a predetermined period of time or longer.

Abstract: The continual engagement support device 600 includes: the performance data storage means 601 that stores performance data including time data indicating the time of performance and sensor data of the time of performance or the surrounding time obtained from the sensors mounted on the portable terminal when the target action is performed via the portable terminal possessed by the target user; the timing detection means 602 that detects the performance timing, which is the timing at which the target user is likely to perform the target action during the predetermined cycle, on the basis of the performance data; and the notification means 603 provides a notification to prompt the performance of the target action at the detected performance timing when the target action is not performed for a predetermined period of time or longer.

Abstract: A state deciding unit 2 makes a decision of double talk when both received signal and transmission signal are in a speech burst, and when the amount of reduction in an echo removal level per prescribed time period exceeds a predetermined threshold, the echo removal level being represented as a level difference between the transmission signal and a residual signal passing through the echo elimination by an adaptive filter 1, and when the amount of changes of the signal level of the received signal is less than a prescribed reference value.

Abstract: A state deciding unit 2 makes a decision of double talk when both received signal and transmission signal are in a speech burst, and when the amount of reduction in an echo removal level per prescribed time period exceeds a predetermined threshold, the echo removal level being represented as a level difference between the transmission signal and a residual signal passing through the echo elimination by an adaptive filter 1, and when the amount of changes of the signal level of the received signal is less than a prescribed reference value.

Abstract: An echo canceller has a high-pass filter 11, a speaker 3, an A/D converter 6, and an echo canceller 7. The high-pass filter 11 suppresses a low-frequency component in a received signal. The speaker 3 outputs an acoustic sound of a low-frequency suppressed received signal passed through the high-pass filter 11. The A/D converter 6 converts an acoustic echo from a microphone 4 into a transmission signal of a digital form. The echo canceller 7 generates a pseudo echo signal based on the low-frequency suppressed received signal passed through the high-pass filter 11, and eliminates the acoustic echo to be inputted to the microphone 4 from the speaker 3 by subtracting the pseudo echo signal from the digital signal converted by the A/D converter 6.

Abstract: The renewal of a group of first tap coefficients is controlled according to a reception signal, a transmission signal and a first error signal. In contrast, the renewal of a group of second tap coefficients is controlled according to only the reception signal. An echo canceled output signal is produced from the group of first tap coefficients in cases where an estimate precision of the group of first tap coefficients is higher than that of the group of second tap coefficients, and an echo canceled output signal is produced from the group of second tap coefficients in cases of the before-initial estimate, the echo path change and the infinite return loss. Accordingly, the echo canceled output signal, in which the echo is canceled, can be reliably produced regardless of the echo path change or the infinite return loss.

Abstract: An echo canceller has a high-pass filter 11, a speaker 3, an A/D converter 6, and an echo canceller 7. The high-pass filter 11 suppresses a low-frequency component in a received signal. The speaker 3 outputs an acoustic sound of a low-frequency suppressed received signal passed through the high-pass filter 11. The A/D converter 6 converts an acoustic echo from a microphone 4 into a transmission signal of a digital form. The echo canceller 7 generates a pseudo echo signal based on the low-frequency suppressed received signal passed through the high-pass filter 11, and eliminates the acoustic echo to be inputted to the microphone 4 from the speaker 3 by subtracting the pseudo echo signal from the digital signal converted by the A/D converter 6.

Abstract: A first pseudo-echo signal approximating an echo of a reception signal superposed on a transmission signal is produced from a delayed reception signal corresponding to a past time according to filer coefficients, and a first residual signal is produced from the first pseudo-echo signal and a delayed transmission signal. Also, the filer coefficients are renewed to reduce the first residual signal. Therefore, when a single-talk state based on a far-end caller's voice of the reception signal is changed to a double-talk state by adding a near-end caller' voice to the transmission signal, the first filter coefficients deteriorate. Also, a second pseudo-echo signal is produced from a reception signal currently received according to the first filter coefficients renewed.

Abstract: A voice communication device has a control CPU for outputting a speaker amplification value corresponding to volume on a speaker which is adjusted by a terminal user by using a volume control unit, and an echo processing unit for reducing an echo equivalent to output voice, which is obtained by amplifying a received input signal demodulated and voice-decoded according to the speaker amplification value output from the control CPU and outputting the received input signal from the speaker and is mixed with a transmitting input signal input through a microphone, according to the speaker amplification value. Therefore, the echo can be properly removed.

Abstract: An echo canceller which estimates an echo gain from a reception signal level and a transmission signal level; calculates logarithmically converted power of a reception signal level; calculates logarithmically converted power of a transmission signal level; estimates an echo gain based on the reception signal level and the transmission signal level; adds the estimated echo gain to the reception signal level to obtain an estimated echo level, calculates logarithmically converted power of a residual signal level; and compares the estimated echo level with the residual signal level after echo cancellation; wherein the echo of the reception signal contained in the transmission signal is canceled when the estimated echo level exceeds the value obtained by adding a margin to the residual signal level.