Abstract: An apparatus for detecting periodicity in time-series data, which is to be processed, includes a specifying unit for designating positions of coordinates in n-dimensional phase space, which are indicated by incoming time-series data and time-series data of the number n-1 that have been already input, arranged in series with the incoming time-series data; an extraction unit for estimating the frequency of emergence in the positions of the coordinates, by taking into consideration a fluctuation in the time-series data, and extracting at least one characteristic point from the coordinates on the basis of the frequency of emergence; a detection unit for checking whether or not the characteristic point at the start point extracted by the extraction unit conforms to the characteristic point at the end point, and detecting periodicity information about the periodicity formed by every characteristic point extracted by the extraction unit; and an output unit for determining the periodicity in the time-series data on t

Abstract: A rhythm pattern generating apparatus is provided having a layered neural network to perform learning with feedback to generate an output pattern signal indicative of a musical sound pattern. The output pattern signal is generated by the layered neural network with feedback in response to a performance operation of a player. The layered neural network generates the output pattern signal indicative of the musical sound pattern based on both an input pattern signal and a weight signal. The output pattern signal is fed back by the feedback circuit to the layered neural network to perform the learning process. A drum pad can be used to provide an input to the rhythm pattern generating apparatus or, specifically, to gate an input pattern selector for selecting input pattern signals. The layered neural network with the feedback can perform the learning process using a back propagation method.