IMPROVED MULTIPULSE ALGORITHM FOR SPEECH CODING BY MEANS OF ADAPTIVE BOLTZMANN ANNEALING

In this paper we will describe an algorithm, based on the Simulated An nealing technique, for solving the problem of multipulse excitation co mputation for speech coding applications. Multiphase excitation modeli ng is described by two set of parameters: pulse locations and amplitud es. The search for optimal pulse locations is the most burdensome task , since there is no closed form solution. Therefore, suboptimal approa ches are used in order to maintain a good effectiveness in the paramet er estimation while keeping the computational load within reasonable b ounds. As compared to the one described in [2], the proposed algorithm achieves an effective improvement in objective performances of more t han 1.5 dB. The price that must be paid for this performance improveme nt is the higher computational cost that this algorithm requires. The objective results of the proposed algorithm will be compared with the results of the algorithm described in [2], and the computational compl exity will be analyzed.