Pitch detection algorithm based on harmonic wavelet transform has been proposed. The method has two major advantages. One is robustness against noise, and the other one is to detect a pitch in a short duration, such as 20ms. However, processing time for the modified correlation method, as a conventional method is 8.6 times faster than that for the harmonic wavelet based algorithm. Due to development of computers, processing time can be reduced easily by using personal computers. In this paper, a new pitch detection algorithm for parallel processing is proposed. The pitch detection algorithm is based on harmonic wavelet transform, and implemented on a parallel processing system. Simulations on a parallel processing system are performed with some node arrangements. As a result, a processing speed for the proposed parallel algorithm is 6.3 times as faster as that for original algorithm.
展开▼
